This may interest you

[san-tech][03411] 2012/01/25：京速コンピュータ「京」と創薬・医療の産学連携セミナー

2012-01-25T00:00:00.001+09:00

Date: Tue, 06 Dec 2011 20:27:20 +0900
--------------------------------------------------
2012年 1月 25日(水) に以下のセミナーが開催されます：

第 2回京速コンピュータ「京」と創薬・医療の産学連携セミナー
‐HPCI計算生命科学推進プログラム‐

日　時：2012年 1月 25日 13:30-17:40
場　所：東京国際フォーラム
参加費：無料 (事前登録)
主　催：大阪大学大学院基礎研究科
(HPCI戦略プログラム分野1「予測する生命科学・医療および創薬基盤」教育プログラム)
詳　細：http://www.biogrid.jp/pdf/bunya1-2.pdf

[Slides/Papers]: Panasas, Object-Based Storage Device, Cray DVS, Berkeley RAID, Sprite OS Project (Log-Structured File Systems), InCast and etc.

2012-01-12T07:47:00.006+09:00

--------------------------------------------------
There are some useful URLs for the following topics:

Panasas Parallel File System (2004, 07, 08)
Panasas System & pNFS (2011)
Cray DVS (Data Virtualization Service) & Panasas
Object-Based Storage Device
Berkeley RAID
Sprite OS Project (Log-Structured File Systems)
InCast
Personal Web Page

==========
Panasas Parallel File System (2004, 07, 08):

"The Panasas ActiveScale Storage Cluster: Delivering Scalable High Bandwidth Storage"
SC '04 Proceedings of the 2004 ACM/IEEE conference on Supercomputing
David Nagle, Denis Serenyi, Abbie Matthews, Panasas Inc.
http://dl.acm.org/citation.cfm?id=1049998
http://dx.doi.org/10.1109/SC.2004.57

"Integrated System Models for Reliable Petascale Storage Systems"
Brent Welch, Panasas, Inc.
Petascale Data Storage Workshop, Supercomputing '07
Paper
http://www.pdsi-scidac.org/SC07/resources/Welch-Integrated%20System%20Models%20for%20Reliable%20Petascale%20Storage%20Systems.pdf
Slides
http://www.pdsi-scidac.org/SC07/resources/Welch%20System%20Models%20SC07.pdf

"Scalable Performance of the Panasas Parallel File System"
USENIX Conference on File and Storage Technologies 2008 (FAST '08)
Brent Welch, Marc Unangst, and Zainul Abbasi, Panasas, Inc.;
Garth Gibson, Panasas, Inc., and Carnegie Mellon University;
Brian Mueller, Jason Small, Jim Zelenka, and Bin Zhou, Panasas, Inc.
http://www.usenix.org/events/fast08/tech/welch.html (Abstract)
http://www.usenix.org/events/fast08/tech/full_papers/welch/welch.pdf

==========
Panasas System & pNFS (2011):

"Panasas Technology"
Brent Welch, Director of Software Architecture, Panasas, Inc.
HPC Advisory Council Switzerland Workshop 2011, March 21 - 23, 2011
http://www.hpcadvisorycouncil.com/events/2011/switzerland_workshop/pdf/Presentations/Day%201/9_2_Panasas.pdf

"pNFS Update: A standard for parallel file systems"
Brent Welch, Director of Software Architecture, Panasas, Inc.
HPC Advisory Council Switzerland Workshop 2011, March 21 - 23, 2011
http://www.hpcadvisorycouncil.com/events/2011/switzerland_workshop/pdf/Presentations/Day%203/8_pNFS.pdf

"Lessons and Predictions from 25 Years of Parallel Data Storage System Development"
Brent Welch, Director of Software Architecture, Panasas, Inc.
Keynote Speach
6th Parallel Data Storage Workshop held in conjunction with Supercomputing '11,
November 13, 201
http://www.pdsi-scidac.org/events/PDSW11/keynote.html
http://www.pdsi-scidac.org/events/PDSW11/resources/slides/PDSW11-Welch-Keynote.pdf

"Panasas: Parallel storage and pNFS updating"
Derek Burke, Director of Partner Sales, Panasas, Inc.
Russian Supercomputing Conference (RSC 2011), Nov. 22, 2011
http://www.rsc2011.ru/images/stories/presentations/RSC%272011_Panasas%20presentation.pdf

"Panasas High-Performance Storage & pNFS Update"
Brent Welch, Director of Software Architecture, Panasas, Inc.
22nd Machine Evaluation Workshop, 29-30 November 2011
http://www.cse.scitech.ac.uk/disco/mew22/presentations/Panasas.pdf

"The Last Bottleneck: How Parallel I/O can improve application performance"
Rex TANAKIT, Director of Industry Solutions, Panasas, Inc.
HPC Advisory Council Stanford University, Dec. 6, 2011
http://www.hpcadvisorycouncil.com/events/2011/Stanford-Workshop/presentations/11_panasas.pdf
Video (16:40, 480p)：
http://www.youtube.com/watch?v=9LCBjvcJS9w

==========
Cray DVS (Data Virtualization Service) & Panasas
"Cray HPC Products & Roadmap"
Geert Wenes, Cray, Inc.
Russian Supercomputing Conference (RSC 2011), Nov. 22, 2011
http://www.rsc2011.ru/images/stories/presentations/RSC%272011_CRAY%20presentation.pdf
Some Cray XE6 Customers Leveraging DVS (34p)

"Introduction to Cray Data Virtualization Service"
S-0005-4002, Published December 2011
Cray, Inc.
http://docs.cray.com/cgi-bin/craydoc.cgi?mode=View;id=S-0005-4002
http://docs.cray.com/books/S-0005-4002/S-0005-4002.pdf

"GPFS on a Cray XT"
Shane Canon, National Energy Research Scientific Computing Center (NERSC)
Cray User Group (CUG) 2009, May 4-7, 2009
http://www.cug.org/5-publications/proceedings_attendee_lists/CUG09CD/S09_Proceedings/pages/authors/01-5Monday/3A-Canon/canon-paper.pdf
http://cug.org/5-publications/proceedings_attendee_lists/CUG09CD/S09_Proceedings/pages/authors/01-5Monday/3A-Canon/canon-slides.pdf
"DVS as a Centralized File System in CLE"
Jason Temple, Swiss National Supercomputing Centre (CSCS)
Cray User Group (CUG) 2009, May 4-7, 2009
http://www.cug.org/5-publications/proceedings_attendee_lists/CUG09CD/S09_Proceedings/pages/authors/01-5Monday/3A-Temple/Temple_Paper.pdf
http://www.cug.org/5-publications/proceedings_attendee_lists/CUG09CD/S09_Proceedings/pages/authors/01-5Monday/3A-Temple/Temple_Presentation.pdf

==========
Object-Based Storage Device:

"Object-based Cluster Storage Systems", Tutorial (104p)
Brent Welch & David Nagle, Panasas, Inc.,
14th NASA Goddard, 23rd IEEE (MSST2006) Conference on Mass Storage Systems and Technologies
http://storageconference.org/2006/Presentations/panasas.dnagle.pdf

"Simplifying Collaboration in the Cloud
WOS and IRODS Data Grid"
Dave Fellinger, Data Direct Networks
27th IEEE (MSST 2011) Symposium on Massive Storage Systems and
Technologies and Co-located Events
http://storageconference.org/2011/Presentations/MSST/5.Fellinger.pdf

IRODS:
Data Grids, Digital Libraries, Persistent Archives, and Real-time Data Systems
https://www.irods.org/

"SCSI Object-Based Storage Device Commands - 2 (OSD-2)"
ANSI/INCITS 458-2011
Information Technology Industry Council(formerly NCITS)
Approved March 15, 2011
American National Standards Institute, Inc.
http://www.techstreet.com/cgi-bin/detail?doc_no=incits|458_2011;product_id=1801667
*) PDF Download: $30.00USD

==========
Berkeley RAID:

"RAID: A Personal Recollection of How Storage Became a System"
Randy H. Katz, University of California, Berkeley
IEEE Annals of the History of Computing, vol. 32, no. 4, Oct.-Dec. 2010
http://www.computer.org/portal/web/csdl/doi/10.1109/MAHC.2010.66

"A Case for Redundant Arrays of Inexpensive Disks (RAID)"
David A. Patterson, Garth A. Gibson and Randy H. Katz
EECS Department, University of California, Berkeley
Technical Report No. UCB/CSD-87-391, December 1987
http://www.eecs.berkeley.edu/Pubs/TechRpts/1987/5853.html
http://www.eecs.berkeley.edu/Pubs/TechRpts/1987/CSD-87-391.pdf

"A case for redundant arrays of inexpensive disks (RAID)"
David A. Patterson, Garth Gibson, Randy H. Katz
Univ. of California, Berkeley
SIGMOD '88 Proceedings of the 1988 ACM SIGMOD international conference on Management of data
http://dl.acm.org/citation.cfm?doid=971701.50214
http://dx.doi.org/10.1145/971701.50214

"Two Papers on RAIDs"
Peter Ming-Chien Chen, Garth A. Gibson, Randy H. Katz, David A. Patterson and
Martin Schulze
EECS Department, University of California, Berkeley
Technical Report No. UCB/CSD-88-479, December 1988
http://www.eecs.berkeley.edu/Pubs/TechRpts/1988/6053.html
http://www.eecs.berkeley.edu/Pubs/TechRpts/1988/CSD-88-479.pdf

"Coding Techniques for Handling Failures in Large Disk Arrays"
Garth A. Gibson, Lisa Hellerstein, Richard M. Karp, Randy H. Katz and
David A. Patterson
EECS Department, University of California, Berkeley
Technical Report No. UCB/CSD-88-477, December 1988
http://www.eecs.berkeley.edu/Pubs/TechRpts/1988/6055.html
http://www.eecs.berkeley.edu/Pubs/TechRpts/1988/CSD-88-477.pdf

"Disk System Architectures for High Performance Computing"
Randy H. Katz, Garth A. Gibson and David A. Patterson
EECS Department, University of California, Berkeley
Technical Report No. UCB/CSD-89-497, March 1989
http://www.eecs.berkeley.edu/Pubs/TechRpts/1989/6166.html
http://www.eecs.berkeley.edu/Pubs/TechRpts/1989/CSD-89-497.pdf

"Failure correction techniques for large disk arrays"
G. A. Gibson,L. Hellerstein, R. M. Karp, D. A. Patterson
Computer Science Division, Electrical Engineering and Computer Sciences,
University of California at Berkeley, Berkeley, CA
ASPLOS-III Proceedings of the third international conference on Architectural
support for programming languages and operating systems, 1989
http://dl.acm.org/citation.cfm?doid=68182.68194
http://dx.doi.org/10.1145/68182.68194

"Redundant Disk Arrays: Reliable, Parallel Secondary Storage"
Garth A. Gibson, Ph.D. Dissertations [advisor: David A. Patterson]
EECS Department, University of California, Berkeley
Technical Report No. UCB/CSD-91-613, 1990
http://www.eecs.berkeley.edu/Pubs/TechRpts/1990/6373.html

"RAID: high-performance, reliable secondary storage"
Peter M. Chen, Univ. of Michigan, Edward K. Lee, DEC Systems Research Center,
Garth A. Gibson, Carnegie Mellon Univ., Randy H. Katz Univ. of California, Berkeley
David A. Patterson, Univ. of California, Berkeley
ACM Computing Surveys (CSUR), Volume 26 Issue 2, June 1994
http://dl.acm.org/citation.cfm?doid=176979.176981
http://dx.doi.org/10.1145/176979.176981

==========
Sprite OS Project (Log-Structured File Systems):

"Beating the I/O Bottleneck: A Case for Log-Structured File Systems"
John K. Ousterhout and Frederick Douglis
EECS Department, University of California, Berkeley
Technical Report No. UCB/CSD-88-467, October 1988
http://www.eecs.berkeley.edu/Pubs/TechRpts/1988/5760.html
http://www.eecs.berkeley.edu/Pubs/TechRpts/1988/CSD-88-467.pdf

"Pseudo Devices: User-Level Extensions to the Sprite File System"
Brent B. Welch and John K. Ousterhout
EECS Department, University of California, Berkeley
Technical Report No. UCB/CSD-88-424, June 1988
http://www.eecs.berkeley.edu/Pubs/TechRpts/1988/6047.html
http://www.eecs.berkeley.edu/Pubs/TechRpts/1988/CSD-88-424.pdf

"Pseudo-File-Systems"
Brent B. Welch and John K. Ousterhout
EECS Department, University of California, Berkeley
Technical Report No. UCB/CSD-89-499, April 1989
http://www.eecs.berkeley.edu/Pubs/TechRpts/1989/6165.html
http://www.eecs.berkeley.edu/Pubs/TechRpts/1989/CSD-89-499.pdf

"Beating the I/O bottleneck: a case for log-structured file systems"
John Ousterhout, Fred Douglis
Computer Science Division, Electrical Engineering and Computer Sciences,
University of California at Berkeley
ACM SIGOPS Operating Systems Review, Volume 23 Issue 1, Jan. 1989
http://dl.acm.org/citation.cfm?doid=65762.65765
http://dx.doi.org/10.1145/65762.65765

"Naming, State Management, and User-Level Extensions in the Sprite Distributed File System"
Brent Ballinger Welch, Ph.D. Dissertations [advisor: John K. Ousterhout]
EECS Department, University of California, Berkeley
Technical Report No. UCB/CSD-90-567, April 1990
http://www.eecs.berkeley.edu/Pubs/TechRpts/1990/6179.html
http://www.eecs.berkeley.edu/Pubs/TechRpts/1990/CSD-90-567.pdf

Sprite papers
http://www.eecs.berkeley.edu/Research/Projects/CS/sprite/sprite.papers.html

Sprite (operating system)
http://en.wikipedia.org/wiki/Sprite_%28operating_system%29

==========
InCast:

"The Panasas ActiveScale Storage Cluster: Delivering Scalable High Bandwidth Storage"
SC '04 Proceedings of the 2004 ACM/IEEE conference on Supercomputing
David Nagle, Denis Serenyi, Abbie Matthews, Panasas Inc.
http://dl.acm.org/citation.cfm?id=1049998
http://dx.doi.org/10.1109/SC.2004.57

"Measurement and Analysis of TCP Throughput Collapse in Cluster-based Storage Systems"
Amar Phanishayee, Elie Krevat, Vijay Vasudevan, David G. Andersen,
Gregory R. Ganger, Garth A. Gibson, and Srinivasan Seshan,
Carnegie Mellon University
USENIX Conference on File and Storage Technologies 2008 (FAST '08)
http://www.usenix.org/events/fast08/tech/phanishayee.html
http://www.usenix.org/events/fast08/tech/full_papers/phanishayee/phanishayee.pdf

"A (In)Cast of Thousands: Scaling Datacenter TCP to Kiloservers and Gigabits."
Vijay Vasudevan, Amar Phanishayee, Hiral Shah, Elie Krevat, David G. Andersen,
Gregory R. Ganger, Garth A. Gibson.
Carnegie Mellon University Parallel Data Lab Technical Report
CMU-PDL-09-101, Feb. 2009.
http://www.pdl.cmu.edu/PDL-FTP/Storage/CMU-PDL-09-101.pdf

==========
Personal Web Page:

Garth Gibson,
Professor,
Computer Science Dept and Department of Electrical and Computer Engineering,
Carnegie Mellon University
Co-Founder and Chief Technology Officer,
Panasas, Inc.
http://www.cs.cmu.edu/~garth/

Brent B Welch,
Director of software architecture at Panasas, Inc.
http://www.beedub.com/

David A. Patterson
Professor
Electrical Engineering and Computer Science Department
University of California, Berkeley
http://www.eecs.berkeley.edu/~pattrsn/

Randy H. Katz
The United Microelectronics Corporation Distinguished Professor
Electrical Engineering and Computer Science Department
University of California, Berkeley
http://bnrg.eecs.berkeley.edu/~randy/

John K. Ousterhout
Professor (Research)
Department of Computer Science
Stanford University
http://www.stanford.edu/~ouster/cgi-bin/home.php

Leslie Lamport
Microsoft Research
My Writings:
http://research.microsoft.com/en-us/um/people/lamport/pubs/pubs.html
*) Paxos

[san-tech][03429] Re: PLFS, a Parallel Log-structured File System, Version2..1, 2011-10-13

2011-12-22T09:41:00.000+09:00

Date: Thu, 22 Dec 2011 09:32:30 +0900
--------------------------------------------------
[san-tech][03422] PLFS, a Parallel Log-structured File System, Version 2.1, 2011-10-13
Date: Fri, 16 Dec 2011 21:02:37 +0900
--------------------------------------------------
ちょっとした驚きです。

"Los Alamos and EMC collaborate to advance high-performance computing"
December 21, 2011
http://www.lanl.gov/news/releases/los_alamos_emc_collaborate_to_advance_high_performance_computing.html

"Los Alamos National Laboratory today announced the signing of
   a new Umbrella CRADA (Cooperative Research and Development Agreement)
   with EMC Corporation."
.....
"This first Project Task Statement (PTS) under the Umbrella CRADA is
   focused on support for the U.S. Department of Energy's Exascale
   Initiative and other data intensive programs. The LANL and EMC
   collaboration for the Exascale initiative is aimed at boosting
   high-performance computing levels to the exaflop - a thousand-times
   faster than current petascale capabilities. The project involves
   design and development of an open-source, extremely scalable
   data-management middleware library called the Parallel Log
   Structured File System (PLFS), which will be used on a range of
   computing platforms from small clusters to the largest supercomputers
   in the world"

EMCが PLFS？？？

[san-tech][03428] Slide: 5th Extremely Large Databases Conference (2011/10/18, 19)

2011-12-21T15:56:00.000+09:00

Date: Wed, 21 Dec 2011 15:45:29 +0900
--------------------------------------------------
2011年 10月 18、19日に開催された

5th Extremely Large Databases Conference
October 18-19, 2011
http://www-conf.slac.stanford.edu/xldb2011/

の講演資料が公開されています。

Conference Program
http://www-conf.slac.stanford.edu/xldb2011/Program.asp

SLAC National Accelerator Laboratoryでの開催ですが、例えば

Reference Cases from Industry:
"Real-time Analytics at Facebook"
   Zheng Shao (Facebook)
"Data Infrastructure at LinkedIn"
   Shirshanka Das (LinkedIn)
"Extreme Analytics at eBay"
   Thomas Fastner (eBay)
"Youtube Data Warehouse"
   Biswapesh Chattopadhyay (Google)

もあります。

[san-tech][03427] Re: International Magnetic Tape Storage Roadmap, Part I: Applications & Systems (2011/11/30)

2011-12-21T15:53:00.000+09:00

Date: Wed, 21 Dec 2011 15:27:31 +0900
--------------------------------------------------
[san-tech][03425] International Magnetic Tape Storage Roadmap,
Part I: Applications & Systems (2011/11/30)
--------------------------------------------------
LTO Generation 8 (最大 32TB) までのロードマップの発表が 2010年 4月 14日
なので、128TBの出荷は・・・

"LTO Program Extends Product Roadmap Through Generation 8"
April 14, 2010
http://ultrium.com/pdf/LTO%20Roadmap%20Extension%20041410.pdf

LTO Roadmap
http://ultrium.com/technology/roadmap.html
http://ultrium.com/technology/LTO_8GenChart_2011.jpg

製品の方は

"LTO Program Announces Availability of Ultrium Format Generation 6 Technology Licenses"
June 20, 2011
http://ultrium.com/pdf/LTO_Gen_6_licenses.pdf
ちなみに Generation 5は
"LTO Program Outlines Upcoming Generation 5 Specifications for Licensing"
January 19, 2010
http://ultrium.com/pdf/LTO-Gen_5_Spec_Release.pdf

Massive Storage Systems and Technology (MSST) 2011
http://storageconference.org/2011/Presentations.html
を眺めていたら

High Availability Tape

セッションがあります。MSSTはセッションが毎年変わるので Tapeは旬な話題なの
でしょう。

ちなみに Keynoteは
"Experience with 30PB of Data from the LHC"
Michael Ernst, Brookhaven National Laboratory and the ATLAS collaboration
http://storageconference.org/2011/Presentations/MSST/1.Ernst.pdf
※BNLは大規模テープサイトです。

CERN ATLAS関係ですが、以下の論文誌が CERN検出器の特集号です：
Journal of Instrumentation
THE CERN LARGE HADRON COLLIDER: ACCELERATOR AND EXPERIMENTS
Volume 3, August 2008
http://iopscience.iop.org/1748-0221/3/08
※Open Accessで、論文 PDFは公開されています。

以前 [san-tech][01999] で紹介した Fulcrum 10Gチップを採用した検出器の
資料のうち SLACは無くなりましたが、ATLAS関連資料は残っています：

Date: Sat, 16 Jan 2010 01:20:43 +0900
> "ATLAS TDAQ upgrade proposal", 2 December 2008
>   http://www.slac.stanford.edu/exp/atlas/upgrade/tdaq/ROD-Upgrade.pdf
> ATLAS TDAQ Upgrade
>   http://www.slac.stanford.edu/exp/atlas/upgrade/tdaq/
> ATLAS Trigger and DAQ
>   http://www.slac.stanford.edu/exp/atlas/tdaq/

[san-tech][03425] International Magnetic Tape Storage Roadmap, Part I: Applications & Systems (2011/11/30)

2011-12-21T10:40:00.000+09:00

Date: Wed, 21 Dec 2011 10:31:52 +0900
--------------------------------------------------
Information Storage Industry Consortiumから、2011年 11月 30日に

International Magnetic Tape Storage Roadmap
Part I: Applications & Systems (60p)
NOVEMBER 2011
http://www.insic.org/news/A&S%20Roadmap.pdf

が公開されました。

International Magnetic Tape Storage Applications & Systems Roadmap
http://www.insic.org/news/A&S%20Roadmap%20article.html

"The 2011 INSIC Magnetic Tape Applications & Systems Roadmap,
released on November 30, 2011,"
.....
"We believe that, thanks to the contributions of these individuals,
the final report is balanced and credible, and delivers an optimistic
outlook for tape storage going forward. However, we also believe
that this optimistic outlook will be fulfilled only if the tape
storage industry continues to invest in the research and advanced
development necessary to provide the future technologies that are
needed.

The companion INSIC Magnetic Tape Technology Roadmap, which will
detail the research and advanced technology developments required
to achieve 128TB (native capacity) tape cartridges and drives
by 2022, is scheduled for limited release in early 2012."

[san-tech][03424] Slide: 22nd Machine Evaluation Workshop (2011/11/29, 30)

2011-12-21T10:37:00.000+09:00

Date: Wed, 21 Dec 2011 10:25:09 +0900
--------------------------------------------------
2011年 11月 29, 30日に開催された

22nd Machine Evaluation Workshop
29-30 November 2011
http://www.cse.scitech.ac.uk/disco/mew22/index.html

の講演資料が公開されています (一部非公開)。

Programme
http://www.cse.scitech.ac.uk/disco/mew22/programme.html

個人的趣味から

"Increasing Use of Ethernet Standards Based Interconnect Within
High Performance Computing Environments"
John Taylor (Gnodal)

"Developments in Tape Storage and Suitability for HPC Environments"
Steve Mackey (Spectra Logic)
※勢いのあるテープメーカーさん

"Panasas High-Performance Storage & pNFS Update"
Brent Welch (Panasas)

"Scale Isn't Free"
Glen Wright (DataDirect Networks)

"Nexenta in the HPC World"
Andy Bennett (NAS UK Ltd)
※pNFS support計画とのことです。

講演資料が公開されているワークショップ：
Data Management
http://www.cse.scitech.ac.uk/disco/mew22/breakout_data.html

[san-tech][03423] "NILFS 2.0.22 was released", Jun 19, 2011 JST.

2011-12-16T21:34:00.000+09:00

Date: Fri, 16 Dec 2011 21:29:38 +0900
--------------------------------------------------
国産 (?) Log Structured File System: NILFSも継続してますね：

NILFS
http://www.nilfs.org/ja/index.html
http://www.nilfs.org/en/index.html

"NILFS utilities 2.0.24 をリリースしました。nilfs-utils 2.0 の
メンテナンスリリースです。GC の不具合と autoreconf の問題の修正を
バックポートしました。
-- 2011年12月5日
"NILFS utilities 2.0.24 was released. A maintenance release of nilfs-utils 2.0
series. This release backported fixes for a GC failure issue and an autoreconf
error."
-- Dec 5, 2011 JST.

"NILFS 2.0.22 をリリースしました．
最近の修正と幾つかのロバストネスの強化機能をスタンドアローンモジュール
にバックポートしました。"
-- 2011年6月19日
"NILFS 2.0.22 was released. backported recent fixes, robustness enhancements
to the standalone module."
-- Jun 19, 2011 JST

[san-tech][03422] PLFS, a Parallel Log-structured File System, Version 2.1, 2011-10-13

2011-12-16T21:09:00.004+09:00

Date: Fri, 16 Dec 2011 21:02:37 +0900
--------------------------------------------------
ちょっとマニアックかもしれませんが、PLFS, a Parallel Log-structured File System
の開発が進行しています：

PLFS: A Checkpoint Filesystem for Parallel Applications
http://institute.lanl.gov/plfs/
"PLFS: A Checkpoint Filesystem for Parallel Applications"
John Bent, Garth Gibson, Gary Grider, Ben McClelland,
Paul Nowoczynski, James Nunez, Milo Polte, Meghan Wingate
SC '09
http://institute.lanl.gov/plfs/plfs.pdf

Abstract
....
"... we have developed a parallel log-structured file system, PLFS,
which is positioned between the applications and the underlying
parallel file system. PLFS remaps an application’s write access
pattern to be optimized for the underlying file system. Through
testing on Panasas ActiveScale Storage System and IBM's General
Parallel File System at Los Alamos National Lab and on Lustre at
Pittsburgh Supercomputer Center, we have seen that this layer of
indirection and reorganization can reduce checkpoint time by up to
several orders of magnitude for several important benchmarks and
real applications."

PLFS Source Code
http://sourceforge.net/projects/plfs/
http://sourceforge.net/projects/plfs/files/

plfs-2.1.tar.gz: 2011-10-13
README.installから 2行：
1) To install PLFS as a FUSE filesystem.
2) To install PLFS as a ROMIO ADIO layer for MPI-IO.

README.posix冒頭：
"The primary goal of PLFS is to support very high bandwidth parallel
IO for HPC large scale parallel applications especially for N-1
write workloads."

"A secondary goal is to be as POSIX compliant as possible without
sacrificing the primary goal. There are several known cases where
PLFS is intentionally not POSIX compliant in order to satisfy the
primary goal:"
(snip)
==================================================
2011/12/17
PLFSは N-1書き込みバンド幅を増加するために、並列ファイルシステムの上で動く
ミドルウェア、元データを PLFSで Log Structuredに再配置して、並列ファイルシステム
への書き出しバンド幅をかせぐ。書き込みバンド幅に対して読み出しは弱いと思われる
ので、例えばチェックポイントファイルとか

"PLFS: A Fast Checkpoint Filesystem"
John Bent, Garth Gibson, Gary Grider, Ben McClelland, Paul Nowoczynski,
James Nunez, Milo Polte, Meghan Wingate
http://www.pdsi-scidac.org/events/PDSW10/resources/posters/plfs-r10.pdf
※Lustreでの比較グラフあり

"PLFS and HDFS: Enabling Parallel Filesystem Semantics In The Cloud"
Milo Polte, Esteban Molina-Estolan, John Bent, Garth Gibson,
Carlos Maltzahn, Maya B. Gokhale, Scott Brandt
http://www.pdsi-scidac.org/events/PDSW10/resources/posters/PLFS-HDFS-svd10.pdf

POSTER SESSION PARTICIPANTS
http://www.pdsi-scidac.org/events/PDSW10/postersession.html
5th Petascale Data Storage Workshop (PDSW)
Supercomputing '10, November 15, 2010
http://www.pdsi-scidac.org/events/PDSW10/index.html

[san-tech][03421] Research Note: Performance Modeling, Benchmarking and Simulation of High Performance Computer Systems (PMBS11)

2011-12-11T22:17:00.000+09:00

Date: Sun, 11 Dec 2011 22:11:39 +0900
--------------------------------------------------
こちらも Supercomputing 2011 (SC '11) ワークショップです。
Research Noteが公開されています：

2nd International Workshop on
Performance Modeling, Benchmarking and Simulation of High Performance
Computer Systems (PMBS11), November 13th 2011
http://www.dcs.warwick.ac.uk/~sdh/pmbs11/PMBS11/Workshop_Schedule.html

講演タイトル：
"Evaluating Interconnect and Virtualization Performance for
High Performance Computing"
"Predictive Modeling and Analysis of OP2 on Distributed Memory
GPU Clusters"
"Optimizing Matrix Transposes Using a POWER7 Cache Model and
Explicit Prefetching"
"BlackjackBench: Portable Hardware Characterization"
"Towards Autotuning by Alternating Communication Methods"
"Modeling Performance through Memory-Stalls"
"A Preliminary Evaluation of the Hardware Acceleration of the Cray
Gemini Interconnect for PGAS Languages and Comparison with MPI"
"Auto-Generation of Communication Benchmark Traces"
"A Method for Automatically Generating Analogue Benchmark Suites using
Low-Level Hardware Metrics"
"Critical Path-Based Thread Placement for NUMA Systems"
"Benchmarking Disk-based Archival Storage Tiers using Appropriate
Archival Workloads"
"Towards Efficient GPU Sharing on Multicore Processors"
"Memory Access Cycle and the Measurement of Memory Systems"

ACM SIGMETRICS
http://www.sigmetrics.org/index.shtml

[san-tech][03420] Paper: 1st Workshop on High-Performance Computing meets Databases at SC '11(2011/11/18)

2011-12-11T22:14:00.000+09:00

Date: Sun, 11 Dec 2011 21:58:04 +0900
--------------------------------------------------
Supercomputing 2011 (SC '11) で開催されていたこちらのワークショップの
ペーパーが公開されています：

1st Workshop on High-Performance Computing meets Databases
November 18, 2011
http://hpcdb.org/program

発表タイトル：
"Toward Efficient Search for Ultrascale Storage Systems"
"Providing fault tolerance in extreme scale parallel applications:
What can the HPC community learn from the Database community?"
"Scientific Data Services: A High-Performance I/O System with
Array Semantics"
"Enabling Automated HPC/Database Deployment via the AppScale Hybrid
Cloud Platform"
"Exploring Provenance in High Performance Scientific Computing"
"Turning the Tide of the Data Deluge"
"Experimenting Lucene Index on HBase in an HPC Environment"
"Large-Scale Continuous Subgraph Queries on Streams"
"Stargazing through a Digital Veil: Managing a Large Scale Sky Survey
using Distributed Databases on HPC Clusters,

[san-tech][03419] "2011 Thermal Guidelines for Liquid Cooled Data Processing Environments", ASHRAE Technical Committee (TC) 9.9

2011-12-11T21:27:00.000+09:00

Date: Sun, 11 Dec 2011 20:46:01 +0900
--------------------------------------------------
ASHRAE (American Society of Heating, Refrigerating and Air-Conditioning
Engineers, Inc.) Technical Committee (TC) 9.9 から以下のホワイトペーパが
公開されてました：

"2011 Thermal Guidelines for Liquid Cooled Data Processing Environments"
Whitepaper prepared by ASHRAE Technical Committee (TC) 9.9
Mission Critical Facilities, Technology Spaces, and Electronic Equipment
2011/09/23
http://tc99.ashraetcs.org/documents/ASHRAE%202011%20Liquid%20Cooling%20Whitepaper.pdf

Executive Summary (Page 2)
"ASHRAE TC 9.9 created the first edition of the "Thermal Guidelines
for Data Processing Environments" in 2004." ...
.....

"TC 9.9 has recently published (May, 2011) a further update described
in a whitepaper available on TC 9.9’s website. The whitepaper
documents expanded data center environmental guidelines by adding
two more envelopes that are wider in temperature and humidity.

However, these guidelines are for air-cooled IT equipment and do not
address water temperatures provided by facilities for supporting liquid
cooled equipment here" ....

"This document describes classes for the temperature ranges of the
facility supply of water to liquid cooled IT equipment. In addition,
this document reinforces some of the information provided in the Liquid
Cooling Guidelines book on the interface between the IT equipment and
infrastructure in support of the liquid cooled IT equipment."

ASHRAE Technical Committee 9.9
http://tc99.ashraetcs.org/

[san-tech][03418] Slide, Paper: 6th Parallel Data Storage Workshop at SC '11 (2011/11/13)

2011-12-11T21:25:00.000+09:00

Date: Sun, 11 Dec 2011 20:27:19 +0900
--------------------------------------------------
Supercomputing 2011 (SC '11) で開催されていた：

6th Parallel Data Storage Workshop
held in conjunction with Supercomputing '11
November 13, 2011
http://www.pdsi-scidac.org/events/PDSW11/

のペーパーと講演資料が上記 Webサイトで公開されています。

ほとんどはタイトルだけですが

Keynote Speaker - Brent Welch (Panasas)
"Lessons and Predictions from 10 Years of Parallel Data Storage System Development"
http://www.pdsi-scidac.org/events/PDSW11/keynote.html
http://www.pdsi-scidac.org/events/PDSW11/resources/slides/PDSW11-Welch-Keynote.pdf
※↑個人的にお勧めです
※スライドのタイトルでは 25 Yearsとなっています。

SESSION 1: PERFORMANCE AND BENCHMARKING
"Robust Benchmarking for Archival Storage Tiers"
"Extending Scalability of Collective IO Through Nessie and Stagin"
"Parallel I/O and the Metadata Wall"

SESSION 2: SECURITY AND USABILITY
"Horus: Fine-Grained Encryption-Based Security for High Performance Petascale Storage"
"Easing the Burdens of HPC File Management"
"The Purge Threat: Scientists' Thoughts on Usability in the Peta-Scale"

SESSION 3: TRANSFORMATIVE APPROACHES
"In-Situ I/O Processing: A Case for Location Flexibility"
"Pattern-Aware File Reorganization in MPI-IO"
"Power Use of Disk Subsystems in Supercomputers"

Short Announcements
"NSF PRObE: A community facility for systems testing at scale"

POSTER SESSION PARTICIPANTS
http://www.pdsi-scidac.org/events/PDSW11/postersession.html

趣味的ネタですが、Paxosアルゴリズムのオリジナル論文等：
My Writings, Leslie Lamport, Microsoft Research
http://research.microsoft.com/en-us/um/people/lamport/pubs/pubs.html

[san-tech][03417] Slide: OpenFabrics Alliance (OFA) at Supercomputing 11 (SC '11)

2011-12-11T13:49:00.000+09:00

Date: Sun, 11 Dec 2011 13:47:35 +0900
--------------------------------------------------
Supercomputing 2011 (SC '11) での、OpenFabrics Alliance (OFA) メンバー
発表資料が公開されました：

SC11
This folder contains the presentations by members at Supercomputing 2011
(SC11) in November, 2011.
https://www.openfabrics.org/resources/document-downloads/presentations/cat_view/57-ofa-documents/23-presentations/39-trade-shows/72-sc11.html
※ブラウザーによってはアクセス出来ない場合があります。

*) System Fabric Works: Demonstrating RDMA Protocols Over a 40 Gb/s WAN
*) Chelsio: How iWARP Helps in Implementing HPC Cloud Computing
*) Mellanox: InfiniBand FDR is Here! Enhancing MPI Performance and
   Scalability via Network Offloading
*) IBTA: InfiniBand Update Addressing New I/O Challenges in HPC, Cloud,
   and Web 2.0 Infrastructures
*) NetApp: Scale Storage from Seismic Depths to Sequoia Heights
*) Microsoft: Overview of RDMA on Windows
*) System Fabric Works: RDMA Over Converged Ethernet
*) Microsoft: Windows HPC Update
*) Software Forge: Writing Application Programs for RDMA Using OFA Software
   - Training Opportunities
*) Benefits of Membership

[san-tech][03416] IBM GPFS Native RAID & USENIX LISA '11

2011-12-09T17:51:00.000+09:00

Date: Fri, 09 Dec 2011 17:46:00 +0900
--------------------------------------------------
現在開催中の USENIX LISA '11 (Dec. 4 - 9, 2011) の Technical Session
INVITED TALKS II, December 8で発表されている

"GPFS Native RAID for 100,000-Disk Petascale Systems"
Veera Deenadhayalan, IBM Almaden Research Center
http://www.usenix.org/events/lisa11/tech/slides/deenadhayalan.pdf

マニュアルが既にあります：

"GPFS Native RAID Administration and Programming Reference"
Version 3 Release 4 (全 142枚)
http://publib.boulder.ibm.com/epubs/pdf/a2313540.pdf

Chapter 1. Introduction (Page 1)
"GPFS Native RAID is a software implementation of storage RAID
technologies within GPFS. Using conventional dual-ported disks
in a JBOD configuration, GPFS Native RAID implements sophisticated
data placement and error correction algorithms to deliver high
levels of storage reliability, availability, and performance.
Standard GPFS file systems are created from the NSDs defined
through GPFS Native RAID."

LISA '11の他のストレージ関係等のペーパー：

"Bringing Up Cielo: Experiences with a Cray XE6 System,
or, Getting Started with Your New 140k Processor System"
Cory Lueninghoener, Los Alamos National Laboratory, et al.
http://www.usenix.org/events/lisa11/tech/full_papers/Lueninghoener.pdf

"Capacity Forecasting in a Backup Storage Environment"
Mark Chamness, EMC
http://www.usenix.org/events/lisa11/tech/full_papers/Chamness.pdf

"Content-aware Load Balancing for Distributed Backup"
Fred Douglis, EMC, et al.
http://www.usenix.org/events/lisa11/tech/full_papers/Douglis.pdf

LISA '11 Technical Sessions
http://www.usenix.org/events/lisa11/tech/

[san-tech][03415] pNFS in Technology Preview, Red Hat Enterprise Linux 6.2

2011-12-09T12:49:00.000+09:00

Date: Fri, 09 Dec 2011 12:41:57 +0900
--------------------------------------------------
先頃発表された Red Hat Enterprise Linux 6.2に pNFS関連が含まれています

Red Hat Enterprise Linux 6.2 - What's New
http://www.redhat.com/f/pdf/RHEL_6_2_features_benefits.pdf

Parallel NFS (pNFS) in Technology Preview
Transfer rates faster than NFS support
*) pNFS is an extension of NFS that provides significantly larger data
   transfer rates compared to the traditional NFS architectures. This
   is achieved by processing the meta-data (a typical bottle neck for
   I/O throughput) separately from the actual data. By separating
   the meta-data and the data paths, pNFS allows clients to access
   storage devices directly and in parallel. pNFS is a part of standard
   NFS version 4.1 specification.
*) Support for pNFS is limited to client-side functionality for file
   layout only.

12.2. pNFS
"A tech preview is available in Red Hat Enterprise Linux 6.2,
introducing experimental client support for the protocol version
NFSv4.1."
http://docs.redhat.com/docs/en-US/Red_Hat_Enterprise_Linux/6/html/Storage_Administration_Guide/ch12s02.html
Chapter 12. Network File System (NFS)
http://docs.redhat.com/docs/en-US/Red_Hat_Enterprise_Linux/6/html/Storage_Administration_Guide/ch-nfs.html
Storage Administration Guide, Red Hat Enterprise Linux 6
http://docs.redhat.com/docs/en-US/Red_Hat_Enterprise_Linux/6/html/Storage_Administration_Guide/index.html

繰り返しますが、pNFS関連の最新情報は
"The Last Bottleneck: How Parallel I/O can improve application performance"
HPC Advisory Council Stanford University, Dec. 6, 2011
Rex TANAKIT, Director of Industry Solutions, Panasas
http://www.hpcadvisorycouncil.com/events/2011/Stanford-Workshop/presentations/11_panasas.pdf
ビデオ (16:40, 480p)：
http://www.youtube.com/watch?v=9LCBjvcJS9w
だと思います。

RFC5661
"Network File System (NFS) Version 4 Minor Version 1 Protocol"
January 2010
http://www.ietf.org/rfc/rfc5661.txt

RFC共同提案者 (Mike Eisler, NetApp) の Blogから
"Red Hat Releases RHEL 6.2 with pNFS support"
December 06, 2011
http://nfsworld.blogspot.com/2011/12/red-hat-releases-rhel-62-with-pnfs.html

"First Anniversary of Red Hat Enterprise Linux 6 Ushered in with Second Wave of
Enhancements and Top Marks in Benchmark Performance Results"
December 6, 2011
http://www.redhat.com/about/news/prarchive/2011/first-anniversary-of-red-hat-enterprise-linux-6
"... today announced the general availability of Red Hat Enterprise Linux 6.2."

[san-tech][03414] TMS introduces the High Availability, High Performance RamSan-720 (12-TB, SLC, 1U), December 6, 2011

2011-12-09T12:07:00.000+09:00

Date: Fri, 09 Dec 2011 12:02:33 +0900
--------------------------------------------------
Twitterで URLは流したのですが、

"TMS introduces the High Availability, High Performance RamSan-720
(12-TB, SLC, 1U) Flash storage system that scales to 500-TB within a single rack"
December 6, 2011
http://www.ramsan.com/company/pressReleases/149

"The RamSan-720 offers 12 TB of Flash capacity, 5 GB/s of bandwidth,
   and no single point of failure. Its proprietary 2D Flash RAID
   technology delivers a new level of High Availability and data
   protection for mission-critical data sets."

RamSan-720
http://ramsan.com/products/rackmount-flash-storage/RamSan-720
RamSan-720 Data Sheet
http://ramsan.com/files/download/798 PDF!

以下 RamSan-720 Data Sheetから (引用順序は個人的趣味)

High Availability
The RamSan-720 has the reliability features required in true enterprise
storage, including:
*) No single point of failure
*) Four layers of data correction
    Chip-level: Error Correcting Code (ECC)
    Module-level: RAID 5 across Flash Chips
    Module-level: Variable Stripe RAID (VSR)
    System-level: RAID 5 across Flash Modules
*) 2D Flash RAID
*) Fully redundant components
*) Hot swappable Flash Modules
*) Low power (typically 300-400 watts)

6 or 12 TB
Two (dual-port 8 Gb/s Fibre Channel or dual-port 40 Gb/s QDR InfiniBand)

Texas Memory Systems Rolls out the RamSan-720 Storage System",
2011/12/05, RichReport,
20:09, 480p
http://www.youtube.com/watch?&v=XBr7C7_EUT0

"Texas Memory Systems gets some enterprise street cred"
High availability flash array here at last
7th December 2011, The Register
http://www.theregister.co.uk/2011/12/07/tms_ramsan_720/

[san-tech][03413] Re: Slide: HPC Advisory Council Stanford Workshop 2011 (2011/12/06, 07)

2011-12-08T19:26:00.000+09:00

Date: Thu, 08 Dec 2011 18:58:05 +0900
--------------------------------------------------
2011年 12月 6、7と開催された：

HPC Advisory Council Stanford Workshop 2011, December 6-7, 2011
http://www.hpcadvisorycouncil.com/events/2011/Stanford-Workshop/index.php

の講演資料が公開されています。

Agenda/Schedule
http://www.hpcadvisorycouncil.com/events/2011/Stanford-Workshop/agenda.php

HPCに偏りますが、
"Parallel I/O from a User's Perspective"
HPC Advisory Council Stanford University,
Dec. 6, 2011
Katie Antypas, Group Leader, NERSC User Services
http://www.hpcadvisorycouncil.com/events/2011/Stanford-Workshop/presentations/14_LBNL.pdf
ビデオ (50:27, 480p)：
http://www.youtube.com/watch?v=4sJIJ7ilxAA
の講演資料はざっと目を通すことをお勧めします (やっぱり偏ってるかな)。

"The Last Bottleneck: How Parallel I/O can improve application performance"
HPC Advisory Council Stanford University,
Dec. 6, 2011
Rex TANAKIT, Director of Industry Solutions, Panasas
http://www.hpcadvisorycouncil.com/events/2011/Stanford-Workshop/presentations/11_panasas.pdf
ビデオ (16:40, 480p)：
http://www.youtube.com/watch?v=9LCBjvcJS9w
は pNFSの現状と、並列ファイルシステムでのアプリケーションパフォーマンス
向上 (16枚目以降) が説明されています (ちょっと宣伝・・・)。

講演ビデオは
RichReport
http://www.youtube.com/user/RichReport
を散策してみて下さい。

[san-tech][03410] Slide: October 2011 Open Networking Summit (2011/10/17 - 19)

2011-12-04T19:22:00.000+09:00

Date: Sun, 04 Dec 2011 19:19:50 +0900
--------------------------------------------------
2011年 10月 17日～19日に開催された

October 2011 Open Networking Summit
http://opennetsummit.org/past_conferences.html

の講演資料等が上記 Webページで公開されています。

Two tutorials
Keynote Session
Power of SDN: Shaping the Future
SDN for Data Center Networks: Technology, Applications, and Markets
SDN for Enterprise Networks: Technology, Applications, Markets
Keynote and ONF Information session
SDN for Service Provider Networks: Technology, Applications, and Markets
Wrap up Panel: Opportunities and Challenges with OpenFlow/SDN in the Market Place

[san-tech][03409] "日立が実証したデータセンタ冷却方式の選定基準が ITU（国際電気通信連合）の国際規格として採用"、2012/12/02

2011-12-04T19:11:00.001+09:00

Date: Sun, 04 Dec 2011 19:07:50 +0900
--------------------------------------------------
株式会社日立製作所と株式会社日立プラントテクノロジーのプレスリリースです：

"日立が実証したデータセンタ冷却方式の選定基準が ITU（国際電気通信連合）
の国際規格として採用"
2011年12月2日
株式会社日立製作所
株式会社日立プラントテクノロジー
http://www.hitachi-pt.co.jp/news/2011/pdf/20111202.pdf

"..... 2010年度(平成22年度)の総務省受託事業として、「電気通信事業者の
   データセンタにおけるサーバ室冷却効率の最適化に係わる実証実験」を実施
   しました。その結果を元に総務省がデータセンタ冷却方式の選定基準に関する
   寄書提案を行ったところ、このたび、世界三大国際標準化機関の一つである
   ITUの国際規格に採用されました。"
.....

"今回、採用された内容は以下のとおりです。
   1.「スペース効率」を考慮すると、高集積のサーバラック
     （例えば5～8kW/ラック以上）を収容するデータセンタでは、
     「天吊り型局所冷却」を選定すべきである。
   2.「エネルギー効率」を考慮すると、外気湿球温度の高い（例えば15℃以上）
      地域に立地するデータセンタでは、「天吊り型局所冷却」を選定すべきで
      ある。外気湿球温度の低い（例えば15℃以下）地域に立地するデータセンタ
      では、「外気冷却」または「気化冷却」を選定すべきである。
   3.「スペース効率」と「エネルギー効率」の両方を考慮すると、温暖な地域
     （例えば湿球温度15℃以上）に立地するデータセンタでは、
     「天吊り型局所冷却」を選定すべきである。"

日立プラントテクノロジー
省エネ空調システム「Ref Assist」(レフアシスト)
http://www.hitachi-pt.co.jp/ref/index.html

関連記事：
"日立のデータセンター冷却方式の選定基準が国際規格に--ITUが採用"
2011年12月02日、ZDNet Japan
http://japan.zdnet.com/datacenter/analysis/35011307/

[san-tech][03407] Slide: Server Design Summit 2011 (2011/11/29, 30)

2011-12-04T15:30:00.000+09:00

Date: Sun, 04 Dec 2011 13:26:44 +0900
--------------------------------------------------
2011年 11月 29、30日に開催された：

Server Design Summit 2011
http://www.serverdesignsummit.com/

講演資料が公開されています。

Get the Server Design Summit Proceedings
http://www.serverdesignsummit.com/cgi-bin/start.cgi/Array_Pages/Entrance_Proceedings.html

[san-tech][03405] Re: 東大京大でスパコンメーカ交代

2011-12-04T15:28:00.000+09:00

Date: Wed, 30 Nov 2011 14:48:57 +0900
--------------------------------------------------
京大次期スーパーコンピュータシステムについて Registerの記事です：

"Cray notches another XE6-Cascades super deal",
Dancing the Japanese two-step,
29th November 2011,
The Register,
http://www.theregister.co.uk/2011/11/29/cray_xe6_cascade_super_kyoto_university/

Appro社のプレスリリースは出てないようですが、
Cray：

"Cray Selected to Provide a Supercomputer to Kyoto University's Academic
Center for Computing and Media Studies"
Nov 29, 2011
http://investors.cray.com/phoenix.zhtml?c=98390&p=irol-newsArticle&ID=1634415

"This is significant for our company on several fronts. Cray is
   the first non-Japanese supercomputer vendor chosen to be a prime
   contractor and supplier to one of Japan's seven major national
   universities, and Kyoto University will give Cray its first order
   for a Cascade system in Asia. This is an exciting moment for us,
   and we are pleased to be working with such a highly respected
   customer."
   Mamoru Nakano, President of Cray Japan

京都大学：
"次期スーパーコンピュータシステムについて"
2011年11月29日 (火曜日) 掲載
http://www.iimc.kyoto-u.ac.jp/ja/services/comp/whatsnew/news/detail/03980.html

次期スーパーコンピュータについて（2012年度）
http://www.iimc.kyoto-u.ac.jp/ja/services/comp/services/coming.html

東京大学：
"次期スーパーコンピュータシステム決定（ピーク性能1.13 PFLOPS）"
2011-11-14
"東京大学情報基盤センターの次期スーパーコンピュータシステム
（ピーク性能1.13 PFLOPS）決定
―高性能・低消費電力システム―"
http://www.itc.u-tokyo.ac.jp/news/2011/20111114j.pdf

"Fujitsu’s PRIMEHPC FX10 with 1.13 PFLOPS starts operation at the University of Tokyo
in April 2012"
November 14th, 2011
Information Technology Center, The University of Tokyo
http://www.itc.u-tokyo.ac.jp/news/2011/20111114e.pdf

"本日より大規模 SMP 並列スーパーコンピュータシステム
（HITACHI SR16000 モデル M1）の正式サービスを開始します。"
2011-11-25
http://www.itc.u-tokyo.ac.jp/news/2011/20111125.pdf

スーパーコンピューティングニュース
東京大学情報基盤センタースーパーコンピューティング部門
http://www.cc.u-tokyo.ac.jp/support/press/news/
Vol.13, No.6（2011年11月）
2011-12-02

[san-tech][03398] "OCZ Technology Acquires UK Design Team from PLX Technology", 2011-10-05

2011-10-07T01:46:00.001+09:00

Date: Thu, 06 Oct 2011 12:23:22 -0400
--------------------------------------------------
"OCZ Technology Acquires UK Design Team from PLX Technology"
Experienced R&D Team and Access to Intellectual Property to Speed Time
to Market and Reduce Development Costs for Company’s Next Generation
SSD Solutions
2011-10-05
http://www.ocztechnology.com/aboutocz/press/2011/453

"... today announced it has signed a definitive agreement to acquire
the UK Design Team and certain assets from PLX Technology, further
strengthening the company’s global research and development team."

"PLX's UK Design Team has built a reputation for designing innovative
and reliable system-on-chip (SOC) solutions and the acquired
engineering team’s expertise provides OCZ with additional resources
for controller design. Through the acquisition of the engineering
team and the license of intellectual property (IP), OCZ will be able
to accelerate solid state drive development, reducing its time
to market for next generation SSD products, while also reducing
development costs."

関連記事
"OCX buys PLX PCIe design smarts operation"
Snaps up UK R&D centre, IP licences and a 40-person research team
6th October 2011, The Register
http://www.theregister.co.uk/2011/10/06/ocz_buys_plx_randd_op/

"Consumer and business solid state drive supplier OCZ is buying
a UK development centre, IP licences, and a 40-person team from
PLX Technology, possibly for a song."
.....
"PLX is a Sunnyvale-based company that designs and builds
semiconductor-based connectivity products both for consumers and
enterprises, which include PCIe switches and 10Gbit Ethernet
transceivers. Its products are aimed at cloud-based network
infrastructures such as high-performance clustering, solid state
drives, remote optical PCI express distribution and Ethernet
home-networking architectures."
.....
"PLX's other R&D centres in Sunnyvale and Bangalore, India, will
remain focused on switching and networking products for the data
centre and cloud services."

[san-tech][03396] "Red Hat to Acquire Gluster", October 4, 2011

2011-10-05T08:27:00.001+09:00

Date: Tue, 04 Oct 2011 19:16:23 -0400
--------------------------------------------------
"Red Hat to Acquire Gluster"
*) Ushers in New Era of Software-Based Open Storage Platforms to
Unify On-Premise and Cloud
*) Adds Scale-Out Management of Unstructured Data to Red Hat Portfolio
with Gluster's Disruptive Open Source Storage Solutions
October 4, 2011
http://www.redhat.com/promo/storage/press-release.html
http://www.gluster.com/2011/10/04/red-hat-to-acquire-gluster/

"... today announced that it has signed a definitive agreement
to acquire Gluster, Inc., a leading provider of scale-out,
open source storage solutions for standardizing the management
of unstructured data. With this acquisition,"

Red Hat to Acquire Gluster
FAQs
http://www.redhat.com/promo/storage/

両社の Blogから
"A Message from the Founders"
October 4th, 2011, abperiasamy
http://www.gluster.com/2011/10/04/a-message-from-the-founders/
"From 'The Red Hat of Storage' to 'The Storage of Red Hat'"
October 4th, 2011, Ben Golub, President and CEO, Gluster, Inc
http://www.gluster.com/2011/10/04/from-the-red-hat-of-storage-to-the-storage-of-red-hat/
Gluster Blog
http://www.gluster.com/blog/

"Red Hat to Acquire Gluster"
October 4, 2011,
Brian Stevens, CTO and VP, Worldwide Engineering, Red Hat
http://www.redhat.com/promo/storage/brian-stevens.html

関連記事：
"Red Hat snatches storage Gluster file system for $136m"
What's inside Pandora's box, in fact
4th October 2011, The Register
http://www.theregister.co.uk/2011/10/04/redhat_buys_gluster/
==================================================
[san-tech][03354] Gluster Announces Apache Hadoop Storage Compatibility,
August 23rd, 2011
Date: Tue, 23 Aug 2011 22:54:10 +0900
[san-tech][03307] GlusterFS OpenStack対応へ, July 27, 2011
Date: Thu, 28 Jul 2011 22:38:37 +0900
[san-tech][03272] SlideCast: Gluster - A Better Way to Do Storage
Date: Sat, 16 Jul 2011 18:54:34 +0900
[san-tech][03202] Brightcove (Online Video Platform) が GlusterFSを採用
Date: Thu, 23 Jun 2011 15:24:22 +0900

[san-tech][03395] Fusion-io: ioDrive2 and ioDrive2 Duo, Oct. 3, 2011

2011-10-03T23:59:00.001+09:00

Date: Mon, 03 Oct 2011 10:55:29 -0400
--------------------------------------------------
Fusion-ioからのプレスリリースです：

"New Storage Class Memory Beats Fusion ioDrive On All Metrics"
Next-Generation Enterprise Flash Product Delivers More Capacity,
Enhanced Enterprise Reliability, Better Performance, All at
a Lower Cost: Introducing the New ioDrive2 from Fusion-io
Oct. 3, 2011
http://www.fusionio.com/press-releases/new-storage-class-memory-beats-fusion-iodrive-on-all-metrics/
http://www.fusionio.com/load/-media-/1g1rrm/docsPress/PR_FIO_ioDrive2_FINAL.pdf

"... today announced the next evolution of its ioMemory platform
for powering application acceleration and data center optimization.
The company that broke performance barriers, brought you the first
enterprise MLC products, dramatically enhanced reliability metrics
for enterprise flash, and redefined server side flash has done it
all over again with the new ioDrive2 and ioDrive2 Duo."
.....
*) System level integration with the latest Fusion-io VSL 3.0 software
subsystem

.....
Pricing and Availability
"The Fusion-io MLC ioDrive2 and ioDrive2 Duo will be offered in
365 GB, 785 GB, 1205 GB and 2.4 TB capacities, beginning in
late November 2011, followed by SLC products in 400 GB, 600 GB
and 1.2 TB capacities, ..."

ioDrive2
http://www.fusionio.com/platforms/iodrive2/
ioDrive2 Data Sheet
http://www.fusionio.com/data-sheets/iodrive2/
http://www.fusionio.com/load/-media-/1g5n3c/docsLibrary/FIO_ioDrive2Datasheets_Rd9.pdf

Virtual Storage Layer (VSL)
http://www.fusionio.com/platforms/vsl
Virtual Storage Layer (VSL) Data sheet
http://www.fusionio.com/data-sheets/virtual-storage-layer/
ioMemory Virtual Storage Layer (VSL)
http://www.fusionio.com/overviews/vsl-technical-overview/
http://www.fusionio.com/load/-media-/1a7jn7/docsLibrary/VSL_Technical_Overview.pdf

Fusion-io Blog
"Dawn Of A New Data Center - Introducing ioDrive2"
10/03/2011
http://www.fusionio.com/blog/dawn-of-a-new-data-center---introducing-iodrive2/

Three "Bolshoi" Supercomputer Simulations of the Evolution of the Universe, Sep. 29, 2011

2011-10-03T08:20:00.002+09:00

--------------------------------------------------
"Three "Bolshoi" Supercomputer Simulations of the Evolution of
the Universe Announced by Authors from University of California,
New Mexico State University"
September 29, 2011
http://hipacc.ucsc.edu/Bolshoi/News.html

"Two research articles describing the most accurate cosmological
simulation of the evolution of the large-scale structure of
the universe yet made" ... "The results calculated by the Bolshoi
simulation - spectacular visualizations of what the universe was
like at time steps 40 million or 80 million years apart - are
being made publicly available to the world’s astronomers and
astrophysicists..."

View the UC-HiPACC Press Release
http://hipacc.ucsc.edu/Bolshoi/BolshoiPR.html
View the UCSC Press Release
http://news.ucsc.edu/2011/09/bolshoi-simulation.html
View the NMSU Press Release
http://newscenter.nmsu.edu/news/article/8043/
View the NASA Ames Press Release
http://www.nasa.gov/centers/ames/news/releases/2011/11-77AR.html

Bolshoi, Cosmological Simulations
http://hipacc.ucsc.edu/Bolshoi/index.html

関連記事：
"NASA Supercomputer Enables Largest Cosmological Simulations"
September 29, 2011, spaceREF
http://www.spaceref.com/news/viewpr.rss.html?pid=34821
"Scientists release most accurate simulation of the universe to date"
September 29, 2011, PHYSORG.COM
http://www.physorg.com/news/2011-09-scientists-accurate-simulation-universe-date.html

[san-tech][03394] Violin Memory: Violin 6000 Series & Violin 3220, Sep. 27, 2011

2011-10-03T06:33:00.001+09:00

Date: Sun, 02 Oct 2011 17:13:40 -0400
--------------------------------------------------
Violin Memoryから新シリーズと新製品の発表がありました：

"VIOLIN MEMORY LAUNCHES POWERFUL NEW PRODUCTS FOR CIOs TO BUILD
THE ALL-SILICON DATA CENTER"
"Rack in a Box" Reliability and Consolidation Advantages Poised to
End the Reign of Mechanical Storage Arrays in the Data Center
September 27, 2011
http://www.violin-memory.com/news/press-releases/6000-2/

"... today introduced the Violin 6000 Series flash Memory Arrays,
the industry's first all-silicon storage systems with
the reliability, performance and economics to be deployed
as mission-critical primary storage."
.....
"Violin also unveiled the new Violin 3220, doubling the storage
capacity of the 3200 Series to 20 terabytes (TB) of SLC per
3U shelf."

Violin 6000 Series
http://www.violin-memory.com/products/6000-flash-memory-array/
※フロントデザインがかなり変わりました

Violin 6616 (6632)
*) Flash Technology: SLC (MLC)
*) Raw Capacity: 16TB (32TB)
*) Useable Capacity (w/ vRAID): 12TB (22TB)
*) Sustained RAIDed R/W IOPS (4K): 1,000,000 (500,000)
*) Bandwidth: 4 GB/sec. (2 GB/sec.)
*) W/R latency: 25/90 μsec (25/140 μsec)

Violin Memory System Software
All Violin Memory Arrays run the same system and management software
*) vSHARE
http://www.violin-memory.com/products/vshare-san-attached/
*) vRAID
http://www.violin-memory.com/technology/flash-technology/
*) vCLUSTER
http://www.violin-memory.com/technology/vcluster-tour/

Violin 3200
http://www.violin-memory.com/products/3200-memory-array/

Violin Memory Document Library
http://www.violin-memory.com/support/document-library/
"Violin 6000 series High Performance Primary Flash Memory Array"
"Violin 3220 series High Performance Flash Memory Array"
※要登録です。

Violin Blogから
http://www.violin-memory.com/blog/

"Flash in the Data Center? Part 1 - Roll your own and Garbage Collection"
September 26, 2011
http://www.violin-memory.com/blog/flash-in-the-data-center-part-1-%E2%80%93-roll-your-own-and-garbage-collection/
"Flash in the Data Center? Part 2 - Why not off the shelf SSDs?"
September 30, 2011
http://www.violin-memory.com/blog/flash-in-the-data-center-part-2/
"Flash in the Data Center - Part 3 - What about PCIe Cards?
September 30, 2011
http://www.violin-memory.com/blog/flash-in-the-data-center-part-3-what-about-pcie-cards/

関連記事：
"Violin Memory Zeros In On Primary Storage Tier with New Flash Offerings"
September 27, 2011, HPCWire
http://www.hpcwire.com/hpcwire/2011-09-27/violin_memory_zeros_in_on_primary_storage_tier_with_new_flash_offerings.html

"Violin strings out fresh flash trio"
Will arrays be music to CIOs' ears?
27th September 2011, The Register
http://www.theregister.co.uk/2011/09/27/violin_3000_6000/
==================================================
[san-tech][03296] Violin Memoryをメタデータに使用して、"GPFS Scans 10 Billion Files in 43 Minutes", July 22, 2011
Date: Fri, 22 Jul 2011 14:47:47 +0900
[san-tech][02811] "Violin Memory Accelerates Network Storage with First Multi-Terabyte NFS Cache", November 16, 2010
Date: Thu, 18 Nov 2010 11:16:27 +0900
[san-tech][02658] Violin Memory: 40TB 3U MLC Flash Storage発表
Date: Tue, 21 Sep 2010 01:14:19 +0900
[san-tech][02382] "Violin Memory Acquires Gear6 Assets"
Date: Thu, 17 Jun 2010 08:15:04 +0900
[san-tech][02334] Violin Memory製品ラインナップを刷新
Date: Thu, 27 May 2010 01:56:46 +0900
[san-tech][02186] Violin Memoryに東芝が出資
Date: Fri, 02 Apr 2010 21:07:50 +0900

[san-tech][03393] QLogic: Highest Density, Most Flexible FC/FCoE Top of Rack (TOR) Switch, Sep 28, 2011

2011-10-03T06:19:00.001+09:00

Date: Sun, 02 Oct 2011 16:14:06 -0400
--------------------------------------------------
QLogicからのプレスリリースです：

"QLogic Universal Access Point Switch Offers Highest Density,
Most Flexible, Converged Solution for Storage and Data Networks"
QLogic Universal Access Point Switches and iSR intelligent Storage
Routers Provide Fabric Freedom to Connect Any Server, Any Network,
Any Storage on Any Switch Port
Sep 28, 2011
http://ir.qlogic.com/phoenix.zhtml?c=85695&p=irol-newsArticle&ID=1610757

プレスリリースからの引用ではなく、以下の製品ページから特長をいくつか

QLogic UA5900 Series
http://www.qlogic.com/Products/Switches/Pages/ConvergedSwitches.aspx
http://www.qlogic.com/Resources/Documents/DataSheets/Switches/ds_UA5900_CNSwitch.pdf

*) Unprecedented port density:
Up to 68 SFP device ports per U of rack space
*) 16/8/4Gb Fibre Channel and 10/1Gb FCoE/Ethernet device port speeds
*) Optional 64Gb Fibre Channel trunking/stacking ports
*) Optional 40Gb Ethernet uplink ports
*) Dual, hot-swappable power supplies
*) Intuitive GUI management and wizard-based setup
*) Single-ASIC design?full line speed, zero contention on all 68 ports
*) Aggregate bandwidth: 2176 Gbps
*) Latency: < 200 nanoseconds
*) Lowest power cost per gigabit of bandwidth

プレスリリースの新 intelligent Storage Router (iSR) は、製品ページは
まだ無いようですが、プレスリリースから
"The QLogic intelligent Storage Router (iSR) will allow cross-protocol
routing for 16Gb Fibre Channel, 10GbE/1GbE iSCSI and 10GbE FCoE."

Intelligent Storage Routers
http://www.qlogic.com/Products/Routers/Pages/IntelligentStorageRouters.aspx

関連プレスリリース：
"QLogic Simplifies Storage and Data Networking With "Fabric Freedom""
Highly Versatile Product Portfolio Delivers Unprecedented Flexibility,
Dramatically Reduces Complexity and Enables Seamless Infrastructure
Migration
Sep 28, 2011
http://ir.qlogic.com/phoenix.zhtml?c=85695&p=irol-newsArticle&ID=1610754
※製品ページはまだのようです。

関連記事：
"Leatherman style multifunctional network device offered"
QLogic delves in trousers, whips out impressive tool
29th September 2011, The Register
http://www.theregister.co.uk/2011/09/29/qlogic_adaptive_convergence/
==================================================
[san-tech][03158] Emulex & JDSU: 16Gb FCデモ (Interop Las Vegas & EMC World 2011)
Date: Wed, 08 Jun 2011 12:13:57 +0900
[san-tech][03063] Brocade 16Gb FC製品群と Brocade Cloudplex発表 (2011/05/03)
Date: Thu, 05 May 2011 19:30:40 +0900
[san-tech][02692] ANSI T11 16Gb/s FC承認
Date: Wed, 29 Sep 2010 21:19:40 +0900

[san-tech][03391] DDN: SFA10Kシリーズ強化 (SFA10K-X & SFA10K-E), 09.20.11

2011-10-02T05:48:00.001+09:00

Date: Sat, 01 Oct 2011 16:44:16 -0400
--------------------------------------------------
DDNが SFA10Kシリーズを強化しました：

"DataDirect Networks (DDN) Delivers Highest Big Data Performance with
SFA10K-X, World's Fastest Integrated Block and File Storage Appliance"
Award-Winning DDN Storage Fusion Architecture (SFA) Technology Embeds
Applications and File Systems While Optimizing Bandwidth, IOPS and
Capacity Utilization for Cloud and Content
09.20.11
http://www.ddn.com/press-releases/2011/datadirect-networks-ddn-delivers-highest-big-data-performance-sfa10k-x-world%E2%80%99s-f
http://www.ddn.com/sites/default/files/10KXPR_FINAL.pdf

"... today unveiled the Storage Fusion Architecture (SFA) 10K-X,
a leading integrated storage appliance that maximizes application
performance while minimizing total cost of ownership for Big Data,
cloud, and content-intensive environments."

SFA10K-X highlights include:
*) Accelerates big data and large content workloads, with read-write
speeds of 17 GB/s, 25 percent faster than the SFA10000
*) Efficiently exploits solid state drive technology to deliver over
840,000 I/Os per second
*) Supports up to 1200 drives (SSD, SATA, SAS) in a single system,
while requiring only 2 floor tiles
*) Capable of housing up to 1.8 petabytes (PB) in a single data center
rack
*) VMWare and VMWare ESX support to accelerate virtual server and
desktop environments

SFA10K-X
http://www.ddn.com/products/sfa10000
http://www.ddn.com/pdfs/SFA10K-X_Datasheet_v7_0911.pdf
"Largest Scalable Storage for Virtualized Environments
Certified with VMware ESX 4.0 and 4.1"
※VMware Certificationは、DDN製品で初？

同時に追加されたシステム (プレスリリースは見あたりませんね)
SFA10K-E
http://www.ddn.com/products/sfa10000e
http://www.ddn.com/pdfs/SFA10K-E_EXAScaler_Datasheet_v6_0911.pdf

解説記事 (目を通す事をお勧めします)：
"DataDirect embulgens arrays for big data"
Hopes to grunt with the big pigs in the Big Data pigpen
26th September 2011, The Register
http://www.theregister.co.uk/2011/09/26/ddn_sfa10k_x/
.....
"No hardware refreshing is mentioned by DataDirect Networks (DDN)
but the company says a lot about the new parallelised and
multi-threaded version of its SFA O/S, with its RAID and cache
engines and data movers. This software is now VMware-certified,
has large capacity support as well as the go-faster changes,
and is available to existing SFA customers at no charge."
....
"There is a special version of the SFA10K-X, an embedded version
called the SFA10K-E. What's embedded in it is a file system,
because this array can run applications, with DDN saying that
its "SFA O/S allows for embedding applications within the SFA10K-E.""

App-running array
"EMC is developing a capability for its VMAX arrays to run
applications. DataDirect has already done this. It says
the SFA10K-E can run applications, and the first ones it is
offering are file systems - specifically Lustre and IBM's GPFS."

[san-tech][03390] NetApp: 55PB (1TB/sec) Storage for LLNL Sequoia Supercomputer (20PF)'s Lustre, Sep. 28, 2011

2011-10-02T05:09:00.001+09:00

Date: Sat, 01 Oct 2011 15:59:39 -0400
--------------------------------------------------
NetApp Eシリーズが、LLNL (Lawrence Livermore National Laboratory)
で構築中の理論性能 20PF Supercomputer: Sequoia (IBM BlueGene/Q) 用
Lustre: 55PB (1TB/sec) のストレージに選ばれました：

"U.S. Department of Energy Selects NetApp as the Storage Foundation
for One of the World's Most Powerful Supercomputers"
September 28, 2011
http://www.netapp.com/us/company/news/news-rel-20110928-990734.html

"... NetApp today announced that the U.S. Department of Energy
(DoE) has selected the company to provide the storage foundation
for its Sequoia supercomputer."
.....
"To store and manage all of this critical data, LLNL will
leverage a solution built on 55PB of NetApp E-Series storage
that will provide over 1TB/sec to the Lustre file system."

NetApp Data Storage Systems
http://www.netapp.com/jp/products/storage-systems/
http://www.netapp.com/us/products/storage-systems/
E-Series Storage Technology
E7900
E5400
E2600

関連記事：
"NetApp Gets First Petascale Supercomputer Win"
September 28, 2011, HPCWire
http://www.hpcwire.com/hpcwire/2011-09-28/netapp_gets_first_petascale_supercomputer_win.html

"NetApp's E-Series storage, which they inherited when the company
purchased LSI's Engenio business, will be used to provide
55 petabytes of disk arrays for the 20-petaflop Sequoia machine."

"NetApp Moves Into HPC Storage with Engenio Acquisition"
March 10, 2011, HPCWire
http://www.hpcwire.com/hpcwire/2011-03-10/netapp_moves_into_hpc_storage_with_engenio_acquisition.html
"Ingenious NetApp buys Engenio"
Boss gets sweet revenge on ex-employer
9th March 2011, The Register
http://www.theregister.co.uk/2011/03/09/netapp_buying_engenio/

関連するかもしれない記事：
"Big Blue hits DataDirect box with spade"
Rolls corpse behind hedge, introduces new NetApp lovely
9th September 2011, The Register
http://www.theregister.co.uk/2011/09/09/ibm_replaces_ddn_with_netapp/

"IBM is killing off its DCS9900 disk array OEM'ed from DataDirect
Networks, and replacing it with a NetApp Engenio box."
==================================================
[san-tech][01483] ジャイアントセコイア (20PF HPC用インフラ)
Date: Sun, 08 Feb 2009 20:03:28 +0900
[san-tech][01532] Re: ジャイアントセコイア (20PF HPC用インフラ)
Date: Tue, 10 Mar 2009 03:49:14 +0900
[san-tech][03040] ASC Sequoia (20PF HPC) は自然換気採用 (PUE: 1.17)
Date: Thu, 14 Apr 2011 21:15:22 +0900
[san-tech][03247] Slide: ScicomP17 (IBM HPC Meeting), May 9-13, 2011
Date: Mon, 04 Jul 2011 12:39:54 +0900> "Sequoia Update"
> Tom Spelce, Development Environment Group, LLNL
> http://spscicomp.org/wordpress/pages/llnl-preparations-for-2011-delivery-of-20-pflop-sequoia-machine/
> http://spscicomp.org/wordpress/wp-content/uploads/2011/05/spelce-SciComp2011-small.pdf
> ※なんと、Sequoiaの Lustreは ZFSベース (QDR InfiniBand SAN) です。
[san-tech][02394] "Livermore Lab Bellies Up to Native ZFS Linux Port",
2010/06/08, insideHPC
Date: Mon, 21 Jun 2010 08:52:22 +0900

[san-tech][03389] "Lustre 2.1 Released", September 30th 2011, Whamcloud Press Release

2011-10-02T04:12:00.001+09:00

Date: Sat, 01 Oct 2011 15:02:10 -0400
--------------------------------------------------
Whamcloudから、Lustre 2.1リリースアナウンスがありました：

"Lustre 2.1 Released"
September 30th 2011, Whamcloud Press Release
http://www.whamcloud.com/news-and-events/lustre-2-1-released/

"Whamcloud, a venture-backed company formed from a worldwide network
of high-performance computing (HPC) storage industry veterans, today
announced the release of Lustre 2.1 with better control over
virtualized storage and wider platform support. It is the first full
release of Lustre with Whamcloud and OpenSFS hosting the master
codeline. Significant in this release, contribution rules have been
greatly simplified to more openly encourage community contributions."

.....
"Several notable new features have been added in Lustre 2.1.
The maximum OST/MDT size has been increased from 16TB to 128TB,
allowing configurations with 3TB drives. The benefit of this is
better control over virtualized storage and was a major user request.
Another main user request, support for RHEL/CentOS 6 server, has
also been added. There were no deprecations compared to the 2.0
support matrix."

OpenSFS
http://www.opensfs.org/

参考：
[san-tech][03388] Slide: European Lustre Workshop 2011 (2011/09/26, 27)
Date: Sat, 01 Oct 2011 14:59:08 -0400> "Lustre Releases"
> Peter Jones, Support Manager, Whamcloud, Inc.
> Whamcloud Lustre Roadmap
http://www.eofs.eu/fileadmin/workshop2011/02_Peter_Jones_Lustre_Releases_-_EOFS_Workshop_Sept_2011_FINAL.pdf

過去の関連プレスリリース・記事：
"Whamcloud Signs Multi-Year Lustre Development and Performance Contract
with OpenSFS"
August 16th 2011, Whamcloud Press Release
http://www.whamcloud.com/news-and-events/whamcloud-signs-multi-year-lustre-development-and-performance-contract-with-opensfs/

"Lustre file system set for spit 'n' polish"
Whamcloud lands deal from OpenSFS
19th August 2011
http://www.theregister.co.uk/2011/08/19/whamcloud_opensfs_lustre/

"Whamcloud, the startup created in July 2010 to continue development
of the open source Lustre supercomputer file system, has secured
a $2.1m contract from OpenSFS to spruce it up with new features
and functions."

[san-tech][03388] Slide: European Lustre Workshop 2011 (2011/09/26, 27)

2011-10-02T04:09:00.001+09:00

Date: Sat, 01 Oct 2011 14:59:08 -0400
--------------------------------------------------
2011年 9月 26, 27日に開催された、

European Lustre Workshop 2011
September 26-27, Paris, Pullman Paris Bercy Hotel
http://www.eofs.eu/index.php?id=5

の講演資料が上記 Webサイトで公開されています。

PDF版 Agenda:
http://www.eofs.eu/fileadmin/workshop2011/agenda_workshop_2011.pdf

例えば：
"Lustre Releases"
Peter Jones, Support Manager, Whamcloud, Inc.
Whamcloud Lustre Roadmap

"Lessons learned from Lustre file system operation"
Roland Laifer, KIT
※4種類の Lustre (2つは HP SFS) や bwGRiD
http://www.scc.kit.edu/produkte/lustre.php

"Administering Lustre 2.0 at CEA"
Stephane Thiell, CEA/DAM
Shine Python Library and Tools for Lustre
http://lustre-shine.sf.net/
ClusterShell scripts to gather Lustre metrics
http://clustershell.sf.net/
NFS-Ganesha FSAL/LUSTRE
http://nfs-ganesha.sf.net/

"RobinHood Policy Engine"
Thomas LEIBOVICI, CEA
Robinhood Policy Engine
http://robinhood.sf.net/
"Robinhood Policy Engine is a multi-purpose tool for managing content
of large filesystems: audit FS content, accounting, remove old unused
files according to admin-defined policies, customizable alerts based
on file properties, backup data to external storage..."

"Visualization of Lustre RPC Traces with Vampir"
Michael Kluge, TU Dresden

"The ClusterStor Engineered Solution for HPC"
Torben Kling Petersen, PhD, Xyratex

"Lustre at exascale"
Eric Barton, CTO, Whamcloud, Inc.

"Lustre in CETA-CIEMAT"
Alfonso Pardo Diaz, CETA-CIEMAT
※構築実例

"Lustre administration - and how it compares to its rivals"
Daniel Kobras, Science+Computing AG
※Lustreと GPFSの比較

[san-tech][03386] "SDSC Announces Scalable, High-Performance Data Storage Cloud", 09.22.11

2011-09-27T08:16:00.005+09:00

Date: Tue, 27 Sep 2011 08:13:57 +0900
--------------------------------------------------
SDSC (San Diego Supercomputer Center) のプレスリリースです：

"SDSC Announces Scalable, High-Performance Data Storage Cloud"
Web-based System Offers High Durability, Security, and Speed
for Diverse User Base
09.22.11
http://www.sdsc.edu/News%20Items/PR092211_sdsccloud.html

"The San Diego Supercomputer Center (SDSC) at the University of
California, San Diego, today announced the launch of what is
believed to be the largest academic-based cloud storage system
in the U.S., specifically designed for researchers, students,
academics, and industry users who require stable, secure,
and cost-effective storage and sharing of digital information,
including extremely large data sets."

.....
"SDSC's new Web-based system is 100% disk-based and interconnected
by high-speed 10 gigabit Ethernet switching technology, providing
extremely fast read and write performance. With an initial
raw capacity of 5.5 petabytes - one petabyte equals one quadrillion
bytes of storage capacity, or the equivalent about 250 billion pages
of text - the SDSC Cloud has sustained read rates of 8 to 10 gigabytes
(GB) per second that will continually improve as more nodes and
storage are added. That's akin to reading all the contents of
a 250GB laptop drive in less than 30 seconds."
.....
""UC San Diego is one of the most data-centric universities in
the country, so our goal was to develop a centralized, scalable data
storage system designed to meet performance, functionality, and
capacity needs of our researchers and partners across the country,
and to evolve and scale with the needs of the scientific community,"
said Dallas Thornton, SDSC's division director of
cyberinfrastructure services. "Developing this resource in-house
atop the OpenStack platform allows for highly-capable and flexible,
yet extremely cost-effective solutions for our researchers.""

OpenStackベースのようです。

Durability and Security
.....
"The SDSC Cloud leverages the infrastructure designed for
a high-performance parallel file system by using two Arista Networks
7508 switches, providing 768 total 10 gigabit (Gb) Ethernet ports
for more than 10Tbit/s of non-blocking, IP-based connectivity.
The switches are configured using multi-chassis link aggregation
(MLAG) for both performance and failover."

詳細は Webページで。
--------------------------------------------------
2011/10/01

"Supercomputing center targets big, fast storage cloud at academics, industry"

September 29, 2011, ars technica

http://arstechnica.com/business/news/2011/09/supercomputing-center-targets-55-petabyte-storage-at-academics-students.ars

"CERN User Story"

October 6, 2011, OpenStack Conference - BOSTON

http://openstackconference2011.sched.org/event/76a3dad1236a5dc342d6d92b40980975

"... This talk reviews the challenges to record and analyse the
25 Petabytes/year produced by the experiments and the investigations
into how OpenStack could help to deliver a more agile computing
infrastructure."

[san-tech][03385] Slide: HEC FSIO 2011 Workshop (2011/0807 - 10)

2011-09-23T18:04:00.001+09:00

Date: Fri, 23 Sep 2011 17:49:34 +0900
--------------------------------------------------
2011年 8月 7 ～ 10日に開催された、

High-End Computing File Systems and I/O (HEC FSIO) 2011 Workshop
Sunday, August 7, 2011 ? Wednesday, August 10, 2011
http://institute.lanl.gov/hec-fsio/conferences/2011/

のプレゼンテーション資料がほぼ揃いました。

HEC FSIO
http://institute.lanl.gov/hec-fsio/

を見て頂くと解りますが、メーカー/ベンダーサイドではなく DOE/NSF予算で
研究・開発をしている側です。尖った発表もありそうです。

"DOD Announces First Strategy for Operating in Cyberspace", July 14, 2011

2011-09-21T21:32:00.001+09:00

"DOD Announces First Strategy for Operating in Cyberspace"

July 14, 2011

http://www.defense.gov/releases/release.aspx?releaseid=14651

"The Department of Defense released today the DoD Strategy for

Operating in Cyberspace (DSOC). It is the first DoD unified strategy

for cyberspace and officially encapsulates a new way forward for

DoD's military, intelligence and business operations."

.....

""The cyber threats we face are urgent, sometimes uncertain and

potentially devastating as adversaries constantly search for

vulnerabilities," said Deputy Secretary of Defense William J. Lynn III.

"Our infrastructure, logistics network and business systems are heavily

computerized. With 15,000 networks and more than seven million

computing devices, DoD continues to be a target in cyberspace for

malicious activity.""

"Department of Defense Strategy for Operating in Cyberspace"

July 2011

http://www.defense.gov/news/d20110714cyber.pdf

[san-tech][03382] OpenStackの大容量ストレージサービス、Swiftの仕組みと使い方、2011/08/05, 09/20, TechTarget

2011-09-21T21:18:00.002+09:00

Date: Wed, 21 Sep 2011 20:40:52 +0900
--------------------------------------------------
NTTデータの伊藤雅典さん連載の

OSSクラウド基盤 OpenStackの全て [全 7回]、TechTarget
http://techtarget.itmedia.co.jp/tt/news/1101/13/news06.html

から、

"OpenStackの大容量ストレージサービス、Swiftの仕組み" [第 6回]
2011年08月05日 09時00分 UPDATE,
伊藤雅典，NTTデータ, TechTarget
http://techtarget.itmedia.co.jp/tt/news/1108/05/news01.html

"OpenStackのオブジェクトストレージサービス「Swift」の特徴、
アーキテクチャ、利用するためのツールを紹介。Swiftは専用
アプリケーションを用いて、大容量データを高速に取り扱うのに
最適化されたサービスである"

"　今回から2回にわたり、OpenStack ObjectStorage（Swift）の全体像と
具体的な利用手順を紹介する。今回はSwiftの特徴、アーキテクチャ、
利用するためのツールなどの全体的な紹介を行い、次回は具体的な
利用手順について解説する。"

"OpenStackの大容量ストレージサービス、Swiftの使い方" [最終回]
2011年09月20日 09時00分 UPDATE
伊藤雅典，NTTデータ, TechTarget
http://techtarget.itmedia.co.jp/tt/news/1109/20/news02.html

"OpenStackの大容量ストレージサービス、Swiftの利用方法について解説
する。インストールや環境設定の仕方、サービスの起動／停止、Swift
コマンドの使い方などを詳しく紹介する。"

"　前回の記事「OpenStackの大容量ストレージサービス、Swiftの仕組み」
で詳しく解説したように、OpenStack ObjectStorage（Swift）とは
Amazon Simple Storage Service（Amazon S3）のようなオブジェクト
ストレージサービスを提供するソフトウェアである。今回はSwiftの具体的
な使い方について説明する。"

※記事を読むには、TechTargetジャパンへの登録 (無料) が必要です。

[san-tech][03381] Intel Hybrid Memory Cube (HMC) に関する Blog, Sep. 15, 2011, Research@Intel

2011-09-21T21:14:00.001+09:00

Date: Wed, 21 Sep 2011 20:22:02 +0900
--------------------------------------------------
IDF 2011でデモンストレーションされた Intel Hybrid Memory Cube (HMC)
についての Blogです：

"Reinventing DRAM with the Hybrid Memory Cube"
September 15, 2011. Bryan Casper, Research@Intel
http://blogs.intel.com/research/2011/09/hmc.php

"Today, Intel CTO Justin Rattner is demonstrating the Hybrid Memory
Cube, the fastest and most efficient Dynamic Random Access Memory
(DRAM) ever built. I want to give you some background on how
and why we collaborated with Micron on this new memory technology."
.....
.....
"The result of this joint research project between Micron Technology
and Intel has been the development of some key achievements. Last year,
Intel designed and demonstrated an I/O prototype that achieved
a record-breaking 1.4 milliwatts per gigabit per second energy
efficiency that was optimized for this hybrid-stacked DRAM application.
The two companies worked together to jointly develop and specify
a high-bandwidth memory architecture and protocol for a prototype
that was designed and manufactured this year by Micron. This
hybrid-stacked DRAM prototype, known as the Hybrid Memory Cube (HMC),
is the world's highest bandwidth DRAM device with sustained transfer
rates of 1 terabit per second (trillion bits per second). On top of that,
it is also the most energy efficient DRAM ever built when measured
in number of bits transferred versus energy consumed. This
groundbreaking prototype has 10 times the bandwidth and 7 times
the energy efficiency than even the most advanced DDR3 memory module
available."

デモンストレーションについては、例えば：

"Intel targets exascale computing with new technologies"
16 September, 2011. ZDNet UK
http://www.zdnet.co.uk/news/emerging-tech/2011/09/16/intel-targets-exascale-computing-with-new-technologies-40093957/

Intel's Hybrid Memory Cube is a type of low-latency Nand flash memory
technology. Photo credit: Jack Clark
http://www.zdnet.co.uk/i/z5/illo/nw/story_graphics/11sept/intel-hybrid-cube-memory-clark.jpg

Date: Wed, 21 Sep 2011 20:55:08 +0900
--------------------------------------------------
"Justin Rattner氏キーノートスピーチ
～メニイコア時代が到来する"
2011年 9月 20日,【IDF 2011レポート】, PC Watch
http://pc.watch.impress.co.jp/docs/news/event/20110920_478686.html

の最後に、

●超広帯域メモリを実現するHMC技術

があります。

キーノートスピーチ Webcast (3件) はこちらで公開されています：
IDF Keynotes
http://www.intel.com/idf/keynote-speakers/index.htm
==================================================
[san-tech][03160] "Open-Silicon and Micron Align to Deliver Next-Generation Memory Technology", June 7, 2011
Date: Wed, 08 Jun 2011 19:32:01 +0900

[san-tech][03379] Cache IQ：新興 NAS加速化キャッシュプラットフォーム, Sep. 19, 2011

2011-09-21T21:05:00.001+09:00

Date: Wed, 21 Sep 2011 15:31:39 +0900
--------------------------------------------------
Cache IQ社が少しだけ製品アナウンスをしました：

"Cache IQ Announces Industry's First Smart NAS Acceleration Solution"
RapidCache dramatically increases application performance and tames
the out of control growth of network attached storage
September 19, 2011
http://www.cacheiq.com/about/company-news/press-releases-2/

"Cache IQ, an enterprise storage solutions company, announced today
the limited release of their RapidCache solution. This intelligent,
high-performance network attached storage (NAS) caching appliance
delivers enhanced performance to I/O intensive applications while
relieving overburdened storage subsystems. RapidCache identifies
applications' active data sets, and then uses a combination of
DRAM and SSDs to dramatically improve performance."

RapidCache
http://www.cacheiq.com/products/rapid-cache/

RapidCache Datasheet
http://www.cacheiq.com/wp-content/uploads/2011/03/RapidCache-Data-Sheet.pdf
RapidCache Technical Overview, White Paper
http://www.cacheiq.com/wp-content/uploads/2011/05/Cache-IQ-RapidCache-Technical-Overview-05-18-2011.pdf
※White Paperを見ないと構成は解りません。

Avere Systems, Inc.
http://www.averesys.com/
A-3 Architecture
http://www.averesys.com/Products_AvereOS.aspx
※ここの仕組みも良く解ってません。

とのざっと見た違いは、RapidCacheは "Flow Director" と "Data Server"
の　2機種構成です。Data Serverの搭載メディア (D-RAM/SSD)、接続台数は
自由なようです (実運用中の増設可能かは解りません)。
Data Server (Virtual Tier) の自由度を強調している感じがしますが、
(将来のメディア候補は PCIe Flash, iSCSI, FCoE等)、設置台数を含めて、
設計が重要ですね。当然、キャッシュヒット率等の管理ツールの完成度が
重要なカギとなりそうです。

う～ん、Intelligent High Performance NAS Cache Platform

Date: Wed, 21 Sep 2011 15:44:30 +0900
--------------------------------------------------
Avere Systems, Inc.に関する The Registerの最新記事です：

"Avere menaces NetApp with accelerators"
Promises satisfaction without short-stroking
19th September 2011, The Register
http://www.theregister.co.uk/2011/09/19/avere_threatens_netapp/

"Avere says it is capturing business that would have gone to NetApp
because its clustered accelerators cost less than NetApp storage
upgrades, and run faster."
.....
"Up to six FXT nodes can be clustered together and any NFS-using
filers can be accelerated by Avere. It is focusing on NetApp
and EMC because they are the big fish in its sea. El Reg
was briefed by Avere CEO Ron Bianchini on a couple of recent wins
against NetApp."

以下導入例の紹介で、最後に

"Bianchini says Avere FXTs have been used in front of BlueArc
filers too. But, we said, these are hardware-accelerated and
go fast. "No it doesn't," Bianchini replied: "It's one box and
eventually the box runs out of gas... We sit in front of BlueArc,
Isilon, whatever.""
==================================================
[san-tech][02158] アリス・イン・ワンダーランドは Avereで
Date: Sun, 21 Mar 2010 16:16:22 +0900

[san-tech][03378] ScaleMP Announces Strategic Collaboration with AMD, Sep. 20, 2011

2011-09-21T21:00:00.001+09:00

Date: Wed, 21 Sep 2011 14:59:44 +0900
--------------------------------------------------
ScaleMPから AMD公式サポートのアナウンスがありました：

"ScaleMP Announces Strategic Collaboration with AMD:
vSMP Foundation for AMD-Based Servers"
Users can scale servers based on the AMD Opteron platform beyond
four processors and 512GB RAM using vSMP Foundation
September 20, 2011
http://www.scalemp.com/scalemp-announces-strategic-collaboration-with-amd-vsmp-foundation-for-amd-based-servers

"ScaleMP, ..., today announced a strategic collaboration with AMD
for vSMP Foundation to support AMD Opteron processor-based servers.
The collaboration enables users to scale AMD-based servers beyond
four processors and create AMD Opteron processor-based systems
with up to 512 processors and 64TB of memory."

Product Availability
"The upcoming release of vSMP Foundation will support aggregating
up to 128 AMD processor-based servers, providing customers with
up to 64TB RAM and 512 sockets, offering thousands of cores.
A limited release of vSMP Foundation will be available to customers
starting October 1, 2011, followed by a general availability
November 21, 2011."

関連記事：
"ScaleMP (finally) glues together 128 Opteron servers"
The 8,192 core, 64TB AMD behemoth
20th September 2011, The Register
http://www.theregister.co.uk/2011/09/20/scalemp_supports_amd_opterons/
.....
"This version of vSMP is now supported on Opteron-based servers,
not just those based on Intel Xeons. ScaleMP is supporting nodes
based on either the current 8-core and 12-core "Magny-Cours"
Opteron 6100s and the 12-core and 16-core Opteron 6200s.
The virtual machine manager at the heart of vSMP can currently
scales to 128 nodes. Depending on the cores per chip and
the generation you use, you can have from 2,048 to 8,192 cores
in a single image."

ScaleMP関連のもう一つ重要なプレスリリース：
"ScaleMP Provides Fujitsu with vSMP Foundation Hardware Certification"
ScaleMP vSMP Foundation and PRIMERGY x86 Servers Combine
To Create a Powerful SMP Solution for HPC Systems
September 15, 2011
http://www.scalemp.com/scalemp-provides-fujitsu-with-vsmp-foundation-hardware-certification

"ScaleMP社の「vSMP Foundation」とPCサーバ「PRIMERGY」を組み合わせた
HPC向けSMPソリューションを提供"
「PRIMERGY RX300 S6」「PRIMERGY RX600 S6」で「vSMP Foundation」の
ハードウェア認証を取得
2011年9月13日
http://pr.fujitsu.com/jp/news/2011/09/13.html

Certified Hardware and OS, ScaleMP
http://www.scalemp.com/spec
※Backplane Infiniband HCAは Mellanoxだけ・・・
==================================================
[san-tech][03174] ScaleMP大規模導入事例 (768コア、6TB、ポーランド)
Date: Thu, 16 Jun 2011 21:55:32 +0900
[san-tech][03322] ScaleMP: STREAM’s Triad function was just over 2.25 TB/s,

August 01, 2011
Date: Tue, 02 Aug 2011 02:55:18 +0900

[san-tech][03377] デル、最高45℃での使用も保証で外気冷却を支援、2011/08/24、＠IT

2011-09-19T19:52:00.001+09:00

Date: Mon, 19 Sep 2011 19:40:02 +0900
--------------------------------------------------
デル株式会社のプレスリリースリストには見当たりませんが、

"複数の既存サーバやストレージを対象に
デル、最高45℃での使用も保証で外気冷却を支援"
2011/08/24、＠IT
http://www.atmarkit.co.jp/news/201108/24/dell.html

"　デルは8月24日、事業者や企業のデータセンターにおける外気冷却への
取り組みを支援するため、同社の法人向けIT機器の保証でカバーする
温度範囲を拡大したと日本で発表した。1機種に限定したものでなく、
同社の複数の既存サーバ、ストレージ、スイッチなどに適用される。

　具体的には、年間の合計稼働時間のうち、10％までを室温5～40℃、
湿度5～85％、1％までを室温-5～45℃、湿度5～90％、そして残りの時間を
10～35℃、湿度10～80％の範囲で動かす場合、これがデルの製品保証範囲に
含まれるというもの。常時45℃の稼働を保証するというものではない。
目的が外気冷却の活用支援だからだ。"

複数の既存IT機器を対象製品として認定した：
http://www.atmarkit.co.jp/news/201108/24/l_dell02.jpg

※積算温度モニタシステムみたいなものが必要になりそう。

Dell (US) のプレスリリース：

"Dell Leads Shift to "Chiller-Less" Data Centers with Fresh Air Technology"
7/28/2011
http://content.dell.com/us/en/corp/d/press-releases/2011-07-28-fresh-air-initiative.aspx
.....
"Trends in data center design place an increased emphasis on
the efficiency and cost benefits that can be achieved with higher
operational temperatures. In addition to ASHRAE's updated standard,
The Institute for Energy, a European guidance body, has issued
the "European Union Code of Conduct for Data Centres," which
challenges IT manufacturers to increase their allowable environmental
limits and sets a goal of enabling chiller-less facility designs."

"European Union Code of Conduct for Data Centres," のリンク先
EU Code of Conduct for Data Centres
http://re.jrc.ec.europa.eu/energyefficiency/html/standby_initiative_data_centers.htm
※ASHRAEに関しては以前伝えた通りで、最後に一部引用します。
.....
"... In line with the new, more stringent ASHRAE A3 and A4 classifications,
Dell systems have been developed for sustained operation at temperature
ranges from minus 23 F (5 C) to 113 F (45 C) and allowable humidity
from 5 percent to 90 percent. This level of design robustness has been
validated by recent tests indicating that the products can tolerate
up to 900 hours of 104 F (40 C) operation per year and up to 90 hours
at 113 F (45 C)"

Dell Technical White Paper
"Chiller-less Facilities: They May Be Closer Than You Think"
June 2011, David Moss, et al.,
http://attachments.wetpaintserv.us/5HcNf7EI-Y_HGVO2PuWj0w411057
※↑PDFファイルです

関連記事
"Dell makes hotter servers for warmer ASHRAE buildings - by Peter Judge"
Aug 2 2011, Green Data Center News
http://www.greendatacenternews.org/articles/292572/dell-makes-hotter-servers-for-warmer-ashrae-buildi/

"Currently 70% of Dell's Enterprise Hardware is ready for running @ 113 F"
AUGUST 23, 2011, Green (Low Carbon) Data Center Blog
http://www.greenm3.com/gdcblog/2011/8/24/currently-70-of-dellrsquos-enterprise-hardware-is-ready-for.html

==================================================
[san-tech][03104] ASHRAE Whitepaper - 2011 Thermal Guidelines公開
Date: Sun, 22 May 2011 19:44:36 +0900> "Expanded Data Center Classes, Guidance Provide More Data Center
> Energy Efficiency Options"
> May 06, 2011
> http://www.ashrae.org/pressroom/detail/more-data-center-energy-efficiency-options
>
> "The major change that is introduced in the whitepaper is the
> addition of two new datacenter classes. "
>
> これは、改訂中 "Thermal Guidelines for Data Processing Environments
> (2004年版)" からの一部を抜き出した Whitepaperで、2011年版の "Thermal
> Guidelines for Data Processing Environments" の発行は年末予定です。
>
>
> "2011 Thermal Guidelines for Data Processing Environments Expanded
> Data Center Classes and Usage Guidance"
> Whitepaper prepared by ASHRAE Technical Committee (TC) 9.9
> http://tc99.ashraetcs.org/documents/ASHRAE%20Whitepaper%20-%202011%20Thermal%20Guidelines%20for%20Data%20Processing%20Environments.pdf
>
> "Thermal Guidelines Whitepaper - Expands allowable temperature
> ranges and provides new environmental metrics to guide the data
> center industry."

[san-tech][02992] ASHRAE：サーバ稼働奨励温度範囲を拡大へ
Date: Tue, 01 Mar 2011 20:52:33 +0900
[san-tech][03010] "Energy Efficiency Guide: Data Center Temperature",
2011/03/10, Data Center Knowledge
Date: Fri, 11 Mar 2011 12:46:00 +0900

[san-tech][03375] Pure Storage: the First All-Flash Enterprise Array, August 23, 2011

2011-09-19T17:51:00.001+09:00

Date: Mon, 19 Sep 2011 16:30:46 +0900
--------------------------------------------------

1ヶ月余り前の発表ですが、All-Flash Arrayストレージシステムメーカー
からのプレスリリースです；

"Pure Storage Breaks the Cost Barrier to Mainstream Flash Adoption
in the Data Center with the First All-Flash Enterprise Array"
All-Flash Array is 10x Faster, 10x More Power and Space Efficient,
and Less Expensive than Traditional Disk-Centric Arrays
August 23, 2011
http://www.purestorage.com/our-company/pr-2011-08-23_2.html

Products Overview
http://www.purestorage.com/products/
http://www.purestorage.com/pdf/Pure_Storage_FlashArray_Datasheet.pdf

FlashArray Hardware Specifications
http://www.purestorage.com/products/pure-storage-flash-array.html

Pure Storage's YouTube Channel
http://www.youtube.com/user/purestorage

関連記事、Blog
"New Samsung-funded Flash disruptor uncloaks"
Storage applecarts tumble like leaves in a gale
23rd August 2011, The Register
http://www.theregister.co.uk/2011/08/23/pure_storage_fa_300/

関連プレスリリース：
"Pure Storage Raises $30M in Series C Financing for All-Flash Enterprise
Array Technology"
Samsung and Redpoint Ventures join Greylock Partners,
Sutter Hill Ventures and Angels in Oversubscribed Round for Disruptive
Flash Storage Startup
August 23, 2011
http://www.purestorage.com/our-company/pr-2011-08-23.html

"Start-Up Places Pure Bet On Flash Memory"
AUGUST 23, 2011, WSJ Blog
http://blogs.wsj.com/digits/2011/08/23/startup-places-pure-bet-on-flash-memory/

"Pure Storage surfaces"
August 23, 2011, RayOnStorage Blog
http://silvertonconsulting.com/blog/2011/08/23/pure-storage-surfaces/

[san-tech][03374] "Kaminario and Fusion-io Collaborate to Deliver Enterprise-class All Solid-State SAN Storage", Sept. 13, 2011

2011-09-19T17:49:00.001+09:00

Date: Mon, 19 Sep 2011 16:21:05 +0900
--------------------------------------------------
Kaminarioという All Solid-state SAN Storage Solutionsメーカーと
Fusion-ioとの協業プレスリリースです
※Kaminarioの新製品プレスリリースも同日です (後述します)。

"Kaminario and Fusion-io Collaborate to Deliver Enterprise-class
All Solid-State SAN Storage"
Kaminario K2 storage embeds Fusion-io technology, addressing
enterprises' I/O- and latency-bound applications
Sept. 13, 2011
http://www.kaminario.com/company/press-releases/Kaminario_and_Fusion-io_Collaborate_to_Deliver_Enterprise-class_All_Solid-State_SAN_Storage
※Fusion-ioからのプレスリリースはありません。

"Kaminario, a leading provider of high performance, all solid-state
SAN storage for accelerating critical business applications,
today announced that it is teaming up with Fusion-io (NYSE: FIO),
a provider of a shared data decentralization platform, to develop
and market the next-generation Kaminario K2 storage, the industry's
first all solid-state SAN storage."
.....
"As part of the companies' collaboration, Kaminario will incorporate
the Fusion ioDrive Duo as a memory accelerator in its all
solid-state SAN storage and its Scale-out Performance Storage
Architecture (SPEAR). Kaminario and Fusion-io will also team up
to bring the Kaminario K2 family of products into joint customer
environments. These combined efforts reinforce the companies'
commitment to remove the barriers enterprises currently face
in obtaining high performance and highly available SAN storage
at a price point that makes sense."

Kaminario K2 storage (新製品) に関するプレスリリース：

"Kaminario Announces the Industry's First Enterprise Grade
All Solid-State SAN Storage with Media Freedom of Choice"
New K2 hybrid delivers the fastest, most reliable and
cost-effective application performance
Sept. 13, 2011
http://www.kaminario.com/company/press-releases/Kaminario_Announces_the_Industrys_First_Enterprise_Grade_All_Solid-State_SAN_Storage_with_Media_Freedom_of_Choice

"..... the new Kaminario K2 family offers a choice of media
(DRAM/Flash) in a single system without compromising on
enterprise-class speed, self-healing high availability or ease of use.
Kaminario will incorporate the Fusion-io ioMemory Flash technology
into its K2 architecture, combining best-in-class solid-state
technology with the industry’s first enterprise-grade, all
solid-state SAN storage solution."

Kaminario K2 All Solid-state SAN Storage Solutions
http://www.kaminario.com/products/K2-Solid-State-SAN-Storage/
http://www.kaminario.com/Kaminario_Datasheet.pdf

Kaminario K2 Models
*)Kaminario K2-H - A hybrid of DRAM and/or Flash SSD media
*)Kaminario K2-D - DRAM-only SSD media
*)Kaminario K2-F - Flash-only SSD media
※各種組み合わせ可能が、Kaminarioの特徴でしょう。

Kaminario All Solid-State SAN Storage Technology
http://www.kaminario.com/products/Technology/
Kaminario Scale-Out Performance Storage Architecture (SPEAR)

Kaminario K2 SAN Storage Data Sheets
http://www.kaminario.com/Resources/Data-Sheets/
Kaminario K2 All Solid-State SAN Storage White Papers
http://www.kaminario.com/Resources/White-Papers/
Kaminario K2 Storage Appliance Customer Case Studies
http://www.kaminario.com/Resources/Case-Studies/

YouTube
KaminarioK2チャンネル
http://www.youtube.com/user/KaminarioK2

関連記事：
"DRAM, Flash in same box named after killer mountain"
But it's OK: Unlikely to kill you, not actually mountain
13th September 2011, The Register
http://www.theregister.co.uk/2011/09/13/kaminario_hybrid_k2/

"Kaminario Releases New 'Hybrid' SSD Appliances"
September 14, 2011, INFOSTORE
http://www.infostor.com/san/kaminario-releases-new-hybrid-ssd-appliances.html

"Kaminario, Anobit add multi-level cell flash options"
14 Sep 2011, SearchStorage.com
http://searchstorage.techtarget.com/news/2240084682/Kaminario-Anobit-add-multi-level-cell-flash-options

Date: Mon, 19 Sep 2011 16:57:10 +0900
--------------------------------------------------
Kaminario社自体は、去年 6月に D-RAMベースのストレージシステムを発表
しています：

"Kaminario K2 High-Performance Enterprise Data Appliance Eliminates
Data Access Bottlenecks and Improves Business Performance"
June 14, 2010
http://www.kaminario.com/company/press-releases/Kaminario_K2_High-Performance_Enterprise_Data_Appliance_Eliminates_Data_Access_Bottlenecks_and_Improves_Business_Performance

"Kaminario climbs performance mountain"
K2 DRAM grid-in-a-box
14th June 2010, The Register
http://www.theregister.co.uk/2010/06/14/kaminario_k2/

[san-tech][03373] Quantum to Expand StorNext Appliance Family, Sep 08, 2011

2011-09-19T17:45:00.001+09:00

Date: Sat, 17 Sep 2011 15:21:17 +0900
--------------------------------------------------

Quantumから StorNextアプライアンスがアナウンスされました：
※StorNext Applianceは初めてでは？

"Quantum to Expand StorNext Appliance Family to Simplify Big Data
File Sharing and Archiving"
New offerings will allow industry-proven StorNext software
to deliver unparalleled data management performance, flexibility
and cost-effectiveness
Sep 08, 2011
http://phx.corporate-ir.net/phoenix.zhtml?c=69905&p=RssLanding&cat=news&id=1604740

"Quantum Corp. ....., today announced it is adding new products to
its StorNext appliance family. The expanded family will use the power
of StorNext and market-leading hardware to offer predictable
high-performance file sharing and archiving in purpose-built
configurations of metadata controllers, expansion appliances and disk,
and archive enabled libraries."

High-performance metadata controller
*)The recently announced StorNext M330 is the first in a series of
metadata controller appliances from Quantum.

Expansion appliances and disk
*)StorNext G300 Scale-Out Gateway Appliance
*)The scalable StorNext QM1200, QS1200 and QD6000 Storage Systems
StorNext QM1200 Storage System will be ideal as fast disk for metadata.
The StorNext QD6000 will scale to over 1 PB of data in a single array
※Expansion appliances and diskの URLsは未だのようです。

Fast, economical archives
*)The StorNext Archive Enabled Libraries can be purchased with as few as
41 activated slots or as many as 5,000 activated slots, protecting
data sets that can range from 60 TB to over 7.5 PB.

StorNext M330 Metadata Controller Appliance
http://www.quantum.com/products/software/stornextM330/index.aspx

Data Sheetから (URL直接指定は不可ですが、↑からダウンロード可能です)
Hardware:
6U total height
2 metadata controllers (one for failover)
1 dedicated metadata array
Client OS support:
Linux, Windows, AIX, HP-UX, Solaris, Mac OS
Hardware reliability:
Dual metadata controllers with automatic failover
Redundant power supplies
Redundant cooling fans
Automated SAS I/O path failover
Hot spare drive in metadata array
4GB mirrored RAID battery backed cache
Connectivity:
Two 8 Gbps FC ports, per controller
Three 1GbE ports with RJ45 connectors

StorNext Archive Enabled Library (AEL)
http://www.quantum.com/products/DataManagement/StorNextAEL/index.aspx
※構成はかなり自由自在のようです

StorNext Family of Appliances for Managing Big Data
http://www.quantum.com/products/DataManagement/index.aspx
※現時点では、StorNext M330と StorNext AEL

StorNext (Software単体)
http://www.quantum.com/products/software/stornext/index.aspx

紹介記事 (プレスリリース＋コメント)
"Quantum to Attack Isilon and Others With Scale-Out NAS"
September 16th, 2011, StorageNewsletter.com
http://www.storagenewsletter.com/news/systems/quantum-stornext-m330

[san-tech][03372] IBM 100 petaOPS Supercomputer出願特許公開、September 8, 2011

2011-09-19T17:39:00.001+09:00

Date: Sat, 17 Sep 2011 14:07:59 +0900
--------------------------------------------------
IBMが 2011年 1月 10日に出願した特許書類が、2011年 9月 8日に公開され
ました (特許成立ではありません)：

Title: MULTI-PETASCALE HIGHLY EFFICIENT PARALLEL SUPERCOMPUTER
United States Patent Application: 20110219208
Kind Code: A1
Inventors: Asaad; Sameh ; et al.
Assignee: International Business Machines Corporation
Publication Date: September 8, 2011
Filed Date: January 10, 2011
http://appft.uspto.gov/netacgi/nph-Parser?Sect1=PTO2&Sect2=HITOFF&p=1&u=%2Fnetahtml%2FPTO%2Fsearch-bool.html&r=1&f=G&l=50&co1=AND&d=PG01&s1=20110219208&OS=20110219208&RS=20110219208
※↑The United States Patent and Trademark Office (USPTO)

Imagesをクリックすると出願書類イメージとなるはずですが、私の場合
ブラウザー (Chrome) との相性問題か、分量の問題かで上手くいきません。
※A4 (Letter?) で、649枚です。460枚程度は図表です。
※後述する民間サイトでも PDFファイルを入手出来ますが、画像 PDFなので
単語検索出来ません。上記 URLはフラット HTMLです (ただし読み難い)。

Abstract
"A Multi-Petascale Highly Efficient Parallel Supercomputer of
100 petaOPS-scale computing, at decreased cost, power and footprint,
and that allows for a maximum packaging density of processing nodes
from an interconnect point of view. The Supercomputer exploits
technological advances in VLSI that enables a computing model where
many processors can be integrated into a single Application Specific
Integrated Circuit (ASIC). Each ASIC computing node comprises
a system-on-chip ASIC utilizing four or more processors integrated
into one die, with each having full access to all system resources
and enabling adaptive partitioning of the processors to functions
such as compute or messaging I/O on an application by application
basis, and preferably, enable adaptive partitioning of functions
in accordance with various algorithmic phases within an application,
or if I/O or other processors are underutilized, then can participate
in computation or communication nodes are interconnected by a five
dimensional torus network with DMA that optimally maximize
the throughput of packet communications between nodes and minimize
latency."

Claims
※48項目

Description
SUMMARY (ほんの一部)
"A novel massively parallel supercomputer capable of achieving 107 petaflop
with up to 8,388,608 cores, or 524,288 nodes, or 512 racks is provided.
It is based upon System-On-a-Chip technology, where each processing node
comprises a single Application Specific Integrated Circuit (ASIC).
The ASIC nodes are interconnected by a five-dimensional torus networks
that optimally maximize packet communications throughput and minimize
latency. The 5-D network includes a DMA (direct memory access) network
interface."

以下に Description項目の大部分とほんの一部の引用を列記しますが、正直
ザッと目を通して感じでは、特許文章というより、BlueGene/Q (出願書類でも
明記) の仕様資料に感じます。また CPU構成等、最新 Supercomputerの教科書
にも良いかもしれません (ただし、かなりのバックグランドが要求されます)。
※これを基に図面を整理して、IBM RedBooksで BleGene/Q解説書として提供
して頂けたらとってもハッピーです。

Description (大部分の項目とほんの一部の引用)
SUMMARY (上の引用はほんの一部だけ)
BRIEF DESCRIPTION OF THE FIGURES
DETAILED DESCRIPTION
"The present invention is directed to a next-generation massively
parallel supercomputer, hereinafter referred to as "BluGene"
or "BluGene/Q"."
Compute ASIC Node
I/O Node
Memory Hierarchy-L1 and L1P (かなりの分量)
Look-Up Engine
Hit Queue
Prefetch Data Array
Prefetch Directory
Look-Up and Stream Comparators
LIST Prefetch Comparators
Automatic Stream Detection, Manual Stream Touch
Prefetch-to-Demand-Fetch Conversion Engine
ECC and Parity
Processing Unit (かなりの分量)
L2 Cache
Memory Synchronization Unit
Levels of Synchronization
Full Sync
Non-Cumulative Barrier
Producer Sync
Generation Change Sync
Producer Generation Change Sync
Consumer Sync
Memory Synchronization Interface Unit
Full Sync
Non-Cumulative Barrier
Producer Sync
Generation Change Sync
Producer Generation Change Sync
Consumer Sync
Local Barrier
Physical Design
Software Interface
Long and Short Running Speculation
L1/L1P Hit Race Condition
Coherence
Speculation Model
Multithreading Model
Speculation IDs
Life Cycle of a Speculation ID
L2 as Point of Coherence
ID Ordering for Speculative Execution
Speculation ID Reclaim
Domains
Transactional Memory
Mode Switching
Memory Consistency
Commit Race Window Handling
Transition to Committed State
Thread ID Counters
Conflict Handling
Conflict Recording
Conflict Detection
TLS/TM/Rollback Management
Global Speculation ID Management
Processor Local Configuration
Larx/stcx
Atomic Operations (かなりの分量)
.....
Miscellaneous Memory-Mapped Devices
System Packaging
System Interconnect-Five Dimensional Torus
Reception
Packet Header and Routing
Interrupt Control
Network Support for System Initiated Checkpoint
Interrupts
Mode Changes
Multiple Tag Domains
Integer Operations
Floating Point Operations
Floating Point (FP) Min and Max
Floating Point ADD
Global Clock
Using Reproducibility to Debug a Multiprocessor System
Aspects Allowing a Multiprocessor System to Offer Reproducibility
Deterministic System Start State
A Single System Clock
System-Wide Phase Alignment
System-Wide Synchronization Events
Reproducibility of Component Execution
Deterministic Chip Interfaces
Precise Stopping of System State
Scanning of System State
Recording the Chronologically Exact Hardware Behavior (かなりの分量)
Cooling
"Blue Gene/Q racks are indirect water cooled."
Power Distribution
Power Management
System Software
Systems Software Overview
Light Weight-Kernel
ここまで、200枚程度

上記解説記事 (非常にコンパクトにまとめてあります)
"IBM Patent Filing Details First 100 PFlop Computer"
September 09, 2011, ConceivablyTech
http://www.conceivablytech.com/9303/products/ibm-patent-filing-details-first-100-pflop-computer

関連記事 (主として BlueGene/Qプロセッサの解説)
"IBM's BlueGene/Q super chip grows 18th core"
It's nice to have a spare
22nd August 2011, The Register
http://www.theregister.co.uk/2011/08/22/ibm_bluegene_q_chip/

"Hot Chips 23 - IBMがスパコンエンジンのBlue Gene/Qチップを発表"
2011/09/08, レポート、マイコミジャーナル
http://journal.mycom.co.jp/articles/2011/09/08/hot_chips23_ibm/index.html

"「京」スパコンシステムの詳細が明らかに"
2011/09/05, レポート、マイコミジャーナル
http://journal.mycom.co.jp/articles/2011/09/05/kei_supercomputer/index.html

SS研HPCフォーラム2011、2011年8月25日(木)
http://www.ssken.gr.jp/MAINSITE/activity/sectionmeeting/sci/2011-1/program.html

"次世代スパコン『京』について"
富士通(株)　追永勇次
プロフィール
http://www.ssken.gr.jp/MAINSITE/activity/sectionmeeting/sci/2011-1/profile_oinaga.pdf
講演要旨(9/2版)
http://www.ssken.gr.jp/MAINSITE/download/newsletter/2011/20110825-sci-1/lecture-5/paper.pdf
プレゼン資料(9/2版)
http://www.ssken.gr.jp/MAINSITE/download/newsletter/2011/20110825-sci-1/lecture-5/ppt.pdf

冒頭で触れた民間サイトですが、私は以下のサービスサイトを使用しています：
FreePatentsOnline
http://www.freepatentsonline.com/
該当申請資料 URL
http://www.freepatentsonline.com/y2011/0219208.html
※USPTOより読みやすいページに工夫されています。

以前は登録しないと PDFファイル入手は不可だったと思いますが、
今では普通にアクセスしても入手可能のようです。

★民間サービスサイトなので、アクセスログ等を解析しているでしょうから、
会社等からのアクセスは、個人的にはお勧めしません (それを言い出したら
USPTOはどうなの？　ですが・・・)。

[san-tech][03371] Broadcom to Acquire NetLogic Microsystems, September 12, 2011

2011-09-19T17:33:00.003+09:00

Date: Sat, 17 Sep 2011 12:48:55 +0900
--------------------------------------------------
既にご存じの方も多いと思いますが：

"Broadcom to Acquire NetLogic Microsystems, Inc., A Leader in
Network Communications Processors"
September 12, 2011
http://www.netlogicmicro.com/News/pr/2011/11-09-12merger.asp
http://investor.broadcom.com/releasedetail.cfm?ReleaseID=604481

"Broadcom Corporation ....., and NetLogic Microsystems, Inc.
(Nasdaq: NETL), a leader in high performance intelligent
semiconductor solutions for next generation networks, today
announced they have entered into a definitive merger agreement.
Under the agreement, NetLogic Microsystems shareholders will
receive $50 per share in a transaction of approximately
$3.7 billion, net of cash assumed."

株式市場分析 Blogによる解説：
"ブロードコムによるネットロジック買収はワイヤレス・インフラ
ストラクチャ市場での商機を狙ったもの"
2011年09月12日、Market Hacker
http://markethack.net/archives/51767420.html

個人的にはこれが興味深かった (過去形は早すぎる？) です：
"NetLogic Microsystems Unleashes Groundbreaking XLPR II,
the World's Most Powerful Multi-Core Communications Processors
with Unparalleled Scalability to 640 NXCPUs"
September 7, 2011
http://www.netlogicmicro.com/News/pr/2011/11-09-07xlpII.asp
.....
"As an aggressive early adopter of the state-of-the-art 28nm
process technology, the XLP II processor family features
groundbreaking innovations that deliver a dramatic
5-7x performance enhancement over the existing generation of
XLP processors, which already offer best-in-class performance
and have set the gold standard in multi-core processing today.
The XLP II processor family is designed to deliver over
100 Gigabits-per-second (Gbps) of network processing
performance per device, and over 800Gbps in a clustered,
fully-coherent system, which is an order-of-magnitude beyond
anything currently available in the market."

[san-tech][03369] 小型のシリコンフォトニクス光源を開発、2011/09/16、富士通

2011-09-19T17:30:00.003+09:00

Date: Fri, 16 Sep 2011 14:44:00 +0900
--------------------------------------------------
日本のメーカーも頑張っています：

"CPU間の大容量光インターコネクトに向けて小型のシリコンフォトニクス光源を開発"
光送受信器の送信部小型化と温度調整機構の不要化により、
大容量光インターコネクトの実現に一歩前進
2011年9月16日
http://pr.fujitsu.com/jp/news/2011/09/16.html

"株式会社富士通研究所は、CPU間での大容量・高速通信を可能とする光による
インターコネクト（データ伝送）を実現するために必要となる、光送受信器用
の小型シリコンフォトニクス光源を開発しました。

従来は、光送受信器に搭載される、光を放つシリコンフォトニクス光源と、
光源から発せられた光に情報を乗せる光変調器に温度変化が起こった場合、
光源の発振波長と光変調器の動作波長にずれが生じることにより、光に情報が
乗らなくなる懸念がありました。そのため常におたがいの動作波長を一致
させるための温度制御が不可欠でした。今回、光源の波長と光変調器の
動作波長を自動的に一致させる機構を新しく取り入れることで、従来技術では
必要であった温度調整機構を不要とすることに成功しました。この技術により、
光送受信器の送信部において小型化および従来比2分の1の低消費電力化が
可能となります。

今回開発した技術により、小型で低消費電力な光送受信器を CPUパッケージに
搭載可能になります。将来のエクサフロップス級スパコンやハイエンドサーバ
などにおけるCPU間の光インターコネクトへの適用により、超高速コンピュータ
の実現が期待されます。"

"本技術の詳細は、9月14日（水曜日）から16日（金曜日）までイギリス・ロンドン
で開催された国際会議「8th Group IV Photonics (GFP 2011)」にて発表いた
しました。"

※英語版 (USA、UK) プレスリリースは未だのようです。

The 8th International Conference on Group IV Photonics
http://www.photonicsconferences.org/GFP2011/index.php
Advance Program (updated 12 August 2011)
http://www.photonicsconferences.org/GFP2011/GFP2011_Advance_Program_8-12-2011.pdf

"Hybrid Laser with Si Ring Resonator and SOA for Temperature Control Free
Operation with Ring Resonator-based Modulator,"
S.-H. Jeong, S. Tanaka, S. Sekiguchi, T. Kurahashi, N. Hatori,
S. Akiyama, T. Usuki, T. Yamamoto and K. Morito,
Fujitsu Laboratories Ltd., Atsugi, Kanagawa, Japan

"Si-wire-based hybrid laser consisting of a microring resonator (MRR)
and a semiconductor optical amplifier is demonstrated.
Novel transmitter scheme potentially makes it possible for a MRR-based
modulator to modulate the signal without temperature control"

[san-tech][03368] 2011/10/28 (金)：インテルソフトウェアカンファレンス東京

2011-09-19T17:28:00.004+09:00

Date: Thu, 15 Sep 2011 18:06:30 +0900
--------------------------------------------------
以下の日程で、インテルソフトウェアカンファレンスが東京で開催されます：

会　期：2011年 10月 28日（金) 10:00～17:10 (9:00受付開始)
会　場：シェラトン都ホテル東京 (東京都港区白金台)
主　催：インテルコーポレーション、インテル株式会社
定　員：400名
参加費：無料（事前登録制）
対象者：並列化のための新しいテクノロジーやツールを使用して、
マルチコア・プロセッサー向けにアプリケーションの並列化および
最適化に取り組むソフトウェア開発者
(セッションで主に使用するプログラミング言語：C/C++/Fortran)
http://www.intelsoftwareconferencejapan.com/tokyo.html

スケジュールの一部
基調講演: あらゆるところで - 並列プログラミングの進歩
セッション 1: インテル Parallel Studio XE の概要とデモ
セッション 2: インテルコンパイラーを利用したインテルR AVX の最適化
セッション 3: インテル Parallel Building Blocks (インテルR PBB)
セッション 4: インテル VTune Amplifier XE ケーススタディー
製品ラインナップやライセンス体系のご紹介

残念ながら、Cilkセッションはありませんが：
[san-tech][03360] Intel Cilk Plus is now available in open-source and for GCC 4.7!,
August 9, 2011

[san-tech][03367] "Koomey's Law: Computing efficiency keeps pace with Moore's Law", SEPTEMBER 13, 2011, InfoWorld

2011-09-19T17:24:00.002+09:00

Date: Thu, 15 Sep 2011 17:26:37 +0900
--------------------------------------------------
"Koomey's Law: Computing efficiency keeps pace with Moore's Law"
Computers have historically doubled in energy efficiency every
18 months, according to new study
SEPTEMBER 13, 2011, InfoWorld
http://www.infoworld.com/t/green-it/koomeys-law-computing-efficiency-keeps-pace-moores-law-172681

Koomey's Lawグラフ (Computations per kWhと年代)：
http://www.infoworld.com/sites/infoworld.com/files/media/image/KoomeysLaw.PNG

元論文 (有償公開です)：
"Implications of Historical Trends in the Electrical Efficiency of Computing"
IEEE Annals of the History of Computing,
July-September 2011 (vol. 33 no. 3)
Jonathan G. Koomey, Stanford University
Stephen Berard, Microsoft
Marla Sanchez, Carnegie Mellon University
Henry Wong, Intel
http://www.computer.org/portal/web/csdl/doi/10.1109/MAHC.2010.28

Abstract:
"The electrical efficiency of computation has doubled roughly every year
and a half for more than six decades, a pace of change comparable to
that for computer performance and electrical efficiency in
the microprocessor era. These efficiency improvements enabled
the creation of laptops, smart phones, wireless sensors, and other mobile
computing devices, with many more such innovations yet to come.
The Web Extra appendix outlines the data and methods used in this study."

[san-tech][03366] HDS Announces Acquisition of BlueArc, September 7, 2011

2011-09-08T10:50:00.004+09:00

Date: Thu, 08 Sep 2011 09:23:03 +0900
--------------------------------------------------

"Hitachi Data Systems Announces Acquisition of BlueArc"
September 7, 2011
http://www.bluearc.com/storage-news/press-releases/110907-Hitachi-Data-Systems-Announces-Acquisition-of-BlueArc.shtml
http://www.hds.com/corporate/press-analyst-center/press-releases/2011/gl110907.html

"Hitachi Data Systems Corporation (HDS), ..... today announced
a significant milestone in its strategy to give customers
seamless access to all data, content and information with
the acquisition of BlueArc Corporation, ..... Building upon
a successful 5 year OEM partnership, Hitachi Data Systems and
BlueArc will give customers the unmatched combination of
Hitachi enterprise-class quality, reliability and support with
innovative, highly scalable, high performance BlueArc network
attached storage (NAS). "

HDS Blog:
"All In The Family"
September 7, 2011,
Claus Mikkelsen, Chief Scientist
http://blogs.hds.com/claus/2011/09/all-in-the-family.html
"BlueArc: A Bountiful Garden of BIG Data"
September 7, 2011,
Michael Hay, Vice President of Product Planning
http://blogs.hds.com/technomusings/2011/09/bluearc-a-bountiful-garden-of-big-data.html
"BlueArc: The Jumbo Carrot"
September 7, 2011,
Ken Wood, Director of Strategy in the Office of Technology and Planning
http://blogs.hds.com/technomusings/2011/09/bluearc-the-jumbo-carrot.html

BlueArc Blog
"When one plus one is truly greater than two"
September 7th, 2011, Shmuel Shottan, Chief Technology Officer
http://www.bluearc.com/blueviews/2011/09/07/when-one-plus-one-is-truly-greater-than-two/

関連記事、Blog
"Hitachi Data Acquires BlueArc in Big Data Deal"
September 7th, 2011, Data Center Knowledge
http://www.datacenterknowledge.com/archives/2011/09/07/hitachi-data-acquires-bluearc-in-big-data-deal/

"HDS Buys BlueArc"
September 7, 2011, The Bigger Truth,
Steve Duplessie, Founder of and Senior Analyst, Enterprise Strategy Group (ESG)
http://www.thebiggertruth.com/2011/09/hds-buys-bluearc/
--------------------------------------------------
2011/09/08

"Hitachi Data Systems gobbles BlueArc"
The end of stand-alone NAS supply
8th September 2011, The Register
http://www.theregister.co.uk/2011/09/08/hds_buys_bluearc/

--------------------------------------------------
2011/09/09

"Squinting Through the HDS BlueArc Deal"
September 7, 2011, Wikibon Blog
http://wikibon.org/blog/squinting-through-the-hds-bluearc-deal/

"Interview with Shmuel Shottan, CTO of BlueArc"
September 8, 2011,
Hu Yoshida, Vice President and Chief Technology Officer, HDS
http://blogs.hds.com/hu/2011/09/interview-with-shmuel-shottan-cto-of-bluearc.html

"HDS buys BlueArc"
September 8, 2011, RayOnStorage Blog
http://silvertonconsulting.com/blog/2011/09/08/hds-buys-bluearc/

[san-tech][03364] Texas Memory: 10 TB of eMLC Flash in 1U, August 23, 2011

2011-08-31T10:06:00.001+09:00

Date: Wed, 31 Aug 2011 09:53:07 +0900
--------------------------------------------------
少し前の発表ですが、

"Texas Memory Systems Brings RamSan High Performance and High Reliability
to eMLC Applications"
August 23, 2011
http://www.ramsan.com/company/pressReleases/146

"The RamSan-810 leverages 33 years of design expertise in record-setting
performance and reliability for the eMLC Flash storage market. It packs
10 Terabytes into a 1U enclosure with 4-GB/s bandwidth and 320K random
IOPS. For read-heavy enterprise applications, the RamSan-810 offers
Tier-0 performance and reliability at Tier-1 prices."

....
"... The 10 TB RamSan-810 has powerful 8-Gb Fibre Channel or
QDR InfiniBand interfaces and dissipates only 250 W."

"Reliability of the RamSan-810 is enhanced by the TMS Series-7 Flash
Controller (FPGA, PowerPC based) that utilizes an advanced error
correction algorithm and patented Variable Stripe Raid (VSR) algorithm.
VSR automatically bypasses failing chips and is only available from
TMS. The RamSan-810 is rated for a 10-year life with a 50-TB/day write
workload. With the extra TMS reliability features found in
the Series-7 Flash Controller, TMS Flash products simply outlast
the competition."

RamSan-810
http://www.ramsan.com/products/rackmount-flash-storage/ramsan-810

Interfaces
Expansion Slots: 2 slots
Fibre Channel Ports per Card: 2 ports/card
Fibre Channel Speed per Port: 8 Gb/s
InfiniBand Channel Width: 4 links
InfiniBand Channels per Card: 2 channels/card
InfiniBand Data Rate Standard: QDR
InfiniBand Speed per Channel: 40 Gb/s

関連記事：
"Part 1: eMLC - It's Not Just Marketing"
August 23, 2011, storagetuning, Jamon Bowen (多分 Texas Memoryの方)
http://storagetuning.wordpress.com/2011/08/23/part-1-emlc-%E2%80%93-it%E2%80%99s-not-just-marketing/

"Slidecast: Texas Memory Systems Steps Up with 10-Year RAID SSDs"
08.23.2011, insideHPC
http://insidehpc.com/2011/08/23/slidecast-texas-memory-systems-steps-up-with-10-year-raid-ssds/
2011/08/20, RichReport, 19:30, 480p
http://www.youtube.com/watch?v=-A9T_gTHHRc

"Texas Memory Systems Enters Primary Storage Game with eMLC Flash Offering"
August 23, 2011, HPCwire
http://www.hpcwire.com/hpcwire/2011-08-23/texas_memory_systems_enters_primary_storage_game_with_emlc_flash_offering.html

"TMS buckles up first MLC flash product"
Suffering from smeary stripe? This cleans up in stride
24th August 2011, The Register
http://www.theregister.co.uk/2011/08/24/tms_ramsan_810/

"The RamSan-810 is priced at half the cost per TB of the 710,
with a 10TB RamSan-810 costing the same as a 5TB 710."

RamSan-710 (SLC Flash storage)
http://www.ramsan.com/products/rackmount-flash-storage/ramsan-710
[san-tech][03146] Texas Memory: 1U, 5TB, 5 GB/s, SLC(東芝製),
8GbFC, QDR IB 発表
Date: Fri, 03 Jun 2011 16:05:22 +0900

"MLC drives, PCIe technology give SSDs run for their money"
26 Aug 2011
http://searchstorage.techtarget.com/news/2240069399/Vendors-increasingly-flash-MLC-drives-and-PCIe-to-lower-costs

[san-tech][03363] Fusion-io ioCache = ioDrive + ioTurbine, Aug. 29, 2011

2011-08-31T09:02:00.002+09:00

Date: Wed, 31 Aug 2011 08:46:49 +0900
--------------------------------------------------
早くも製品化です。

"Fusion-io Unveils ioCache Virtualization Solution"
Fusion's ioTurbine Software and Customized ioMemory Accelerator
Virtualizes Flash to Deliver Affordable Performance for
Virtualizing Data-Intensive Enterprise Applications
Aug. 29, 2011
http://www.fusionio.com/press-releases/fusion-io-unveils-iocache-virtualization-solution/
http://www.fusionio.com/load/-media-/1e8imv/docsPress/PR_FIO_ioCache_8.26.11.pdf
.....
"When deployed in a VMware environment, Fusion's ioTurbine software
dynamically provisions ioMemory capacity and I/O performance across
multiple virtual machines (VMs) by using Fusion's VSL OS Subsystem
to house and serve data for each VM. Tight integration of VSL's
host-based flash management and ioTurbine's cache management
eliminates otherwise redundant mapping functions, maximizing I/O
performance and adding a critical and previously missing piece
to virtualized IT environments - a plentiful supply of scalable,
low-latency I/O."

"Through this approach, Fusion-io unites the benefits of flash and
cache without the overhead costs created by integrating each
technology through traditional architectures using storage
protocols and embedded processors. This solution avoids the need
for expensive, performance-oriented networked storage systems,
SSD arrays and memory appliances that struggle to deliver
performance from the far end of a network, instead allowing
enterprises to use cost optimized shared storage systems."
.....
"Available exclusively in bundles with ioTurbine software,
Fusion's ioCachedelivers 600GB of capacity optimized to support
the majority of virtualized application workloads. "

ioCache
http://www.fusionio.com/products/iocache
http://www.fusionio.com/load/-media-/1eek75/imagesProduct/iocache-product.jpg
※デザインセンスが良い！

ioTurbine
http://www.fusionio.com/products/ioturbine/

関連記事：
"Fusion-io Launches ioCache at VMworld 2011"
AUGUST 30TH 2011, SiliconANGLE
http://siliconangle.com/blog/2011/08/30/fusion-io-launches-iocache-at-vmworld-2011/
==================================================
[san-tech][03326] "Fusion-io to Acquire IO Turbine", Aug. 4, 2011
Date: Fri, 05 Aug 2011 18:21:26 +0900

[san-tech][03118] Slidecast: "IO Turbine Launches Accelio SSD Accelerator

for Virtualization", 05.24.2011, insideHPC,com

Date: Thu, 26 May 2011 14:17:11 +0900

[san-tech][03152] Re: Slidecast: "IO Turbine Launches Accelio SSD Accelerator

for Virtualization", 05.24.2011, insideHPC,com

Date: Sun, 05 Jun 2011 17:30:13 +0900

[san-tech][03362] Citrix: Cloud.comをオープンソース化、8/29/2011

2011-08-30T10:45:00.001+09:00

Date: Tue, 30 Aug 2011 10:37:25 +0900
--------------------------------------------------
Citrix社が、2011年 7月に買収した Cloud.comをオープンソース化しました：

"Citrix Opens Up the Cloud for VMware Customers"
Latest Release of CloudStack Brings the Leading Open Source Cloud
Platform to VMware vSphere and OracleVM Users
8/29/2011
http://www.citrix.com/English/NE/news/news.asp?newsID=2315523

"Citrix Systems today announced a new edition of CloudStack, which
will feature enhanced support for both the VMware vSphere and
Oracle VM hypervisors, enabling VMware and Oracle customers
to manage their virtualized servers as a highly available, scalable
cloud computing environment. Citrix CloudStack is not just
a traditional enterprise server virtualization platform with
cloud management layered on top, it is a production-ready, free
and open source cloud infrastructure platform that helps providers
of all types deploy and manage simple, cost-effective cloud services
that are scalable, secure, and open by design. The new release is
based on the same CloudStack solution that has been used by some of
the world's largest and most successful public clouds to roll out
new cloud services up to 50 times faster and at one-fifth the cost
of alternative solutions(1). This is the first release of
the product since Citrix acquired Cloud.com in July."

.....
Product Availability
"Citrix CloudStack with support for VMware vSphere and bare metal
provisioning is available today as a free download from
CloudStack.org or as a commercially supported solution from Citrix.
Oracle VM support will be available in October, followed by
Microsoft Hyper-V later in Q4. CloudStack can be deployed
on-premise or as a hosted cloud service"

CloudStack
http://www.cloudstack.org/

関連記事：
"Cloud.com goes open source"
Citrix sets cloud builder free
29th August 2011, The Register
http://www.theregister.co.uk/2011/08/29/citrix_cloud_com_update/

"After taking control of the CloudStack cloud management framework
through its acquisition of Cloud.com back in July, Citrix Systems is
now open sourcing the code behind the tool. At the same time, it's
adding support for the provisioning of workloads on additional
hypervisors and, for the first time, on bare-metal machines."

"Citrix Opens Up the Cloud for VMware Customers"
AUGUST 29TH 2011, SiliconANGLE
http://siliconangle.com/blog/2011/08/29/citrix-opens-up-the-cloud-for-vmware-customers/

[san-tech][03361] Re: Slide: Flash Memory Summit 2011 Proceedings (2011/08/09 - 11)

2011-08-29T13:00:00.002+09:00

Date: Mon, 29 Aug 2011 12:51:09 +0900
--------------------------------------------------
[san-tech][03341] Slide: Flash Memory Summit 2011 (2011/08/09 - 11)
Date: Fri, 12 Aug 2011 04:36:15 +0900

--------------------------------------------------

↑で紹介した David Floyerさんの記事が更新されました：

"2011 Flash Memory Summit Enterprise Roundup

Last Update: Aug 28, 2011, Wikibon

Originating Author: David Floyer

http://wikibon.org/wiki/v/2011_Flash_Memory_Summit_Enterprise_Roundup

Contents

1 Sentiment at the Flash Summit

2 Courses for Horses

2.1 Traditional SSD in an Array

2.2 Flash-cache

2.3 Flash-only Arrays

2.4 Data in Server Flash Memory

2.5 Flash as an Extension of Main Memory

3 Conclusions

"The Disruptive Nature of NAND Flash Storage Explained"

AUGUST 26TH 2011, SiliconANGLE

http://siliconangle.com/blog/2011/08/26/the-disruptive-nature-of-nand-flash-storage-explained/

"Wikibon CTO and Flash expert David Floyer discusses the present

state of the fast-growing flash market and why this is becoming

a major disruptor in the storage industry, for software as well

as hardware."

[san-tech][03360] Intel Cilk Plus is now available in open-source and for GCC 4.7!, August 9, 2011

2011-08-29T09:25:00.002+09:00

Date: Mon, 29 Aug 2011 09:15:58 +0900
--------------------------------------------------
Intelから C/C++を拡張した並列言語 Cilkがオープンソースとして公開され
ました。Cilkの発音は "silk" です (以下の MITサイトから)。

Intel Cilk Plus
http://software.intel.com/en-us/articles/intel-cilk-plus/
"Intel Cilk Plus is now available in open-source and for GCC 4.7!"

"Parallelism as a First Class Citizen in C and C++, the time has come."
August 9, 2011, Intel Software Blogs
http://software.intel.com/en-us/blogs/2011/08/09/parallelism-as-a-first-class-citizen-in-c-and-c-the-time-has-come/

ソースツリー
gcc - Revision 178148: /branches/cilkplus
http://gcc.gnu.org/svn/gcc/branches/cilkplus/

gcc, gcc-patches MLから
"Announcing the Port of Intel(r) Cilk (TM) Plus into GCC"
15 Aug 2011 13:30:39 -0700
http://gcc.gnu.org/ml/gcc/2011-08/msg00279.html
http://gcc.gnu.org/ml/gcc-patches/2011-08/msg01290.html

Cilkは元来 MITで開発された Multithreaded parallel computing用言語で、
商業化のために独立した会社：Cilk Arts, Inc.を Intelが買収しました。

The Cilk Project, MIT
http://supertech.csail.mit.edu/cilk/
※こちらは独自に研究活動は継続していて、GPLで公開しています。

"... the commercial Cilk++, which extended the Cilk model to C++ and
introduced "reducer hyperobjects" as an efficient means for resolving
races on no Cilk is an especially effective platform for programming
"irregular" applications such as sparse numerical algorithms,
N-body simulations, graph-theory applications, backtracking search,
and cache-efficient stencil computations."

commercial Cilk++は Cilk Artsで開発されたもので、それが今回 Intel買収
を経て改良されオープンソースとして公開されました。

Cilk
http://en.wikipedia.org/wiki/Cilk

日本語解説書：
"Cilkがやってきた―C/C++プログラマーのための並列プログラミング言語"
http://www.amazon.co.jp/dp/4877832602
著　者：菅原清文 (インテル株式会社、i860コンパイラ開発とか)
単行本：254ページ
出版社：カットシステム (2011/01)
発売日： 2011/01

[san-tech][03359] The SSD Review (SSD情報サイト)

2011-08-29T08:07:00.002+09:00

Date: Mon, 29 Aug 2011 07:56:50 +0900
--------------------------------------------------
SSD情報サイトです：

The SSD Review
The Worlds Dedicated SSD Resource and Review Site
http://thessdreview.com/

About Us
http://thessdreview.com/background/
"The SSD Review was officially launched and available to the public
on March 28, 2010."

"The roots of the site can be traced back to August 29, 2007 when
the Publisher, Les Tokar, published the first review of a solid
state drive on the internet with his review of the Sandisk U5000
32Gb SLC SSD."
.....
"All original articles are published here at The SSD Review
in order to assist in an understanding of how fast we are
progressing in the world of SSD storage."

Forums
http://thessdreview.com/Forums/
Register
http://thessdreview.com/register-now/

[san-tech][03358] Panasas Webinar: "pNFS-Are We There Yet?" (Aug. 24, 2011)

2011-08-28T22:10:00.002+09:00

Date: Sun, 28 Aug 2011 22:02:14 +0900
--------------------------------------------------
2011年 8月 24日の Panasasの Webinar: "pNFS-Are We There Yet?" の
ビデオが公開されています：

WEBINAR: PNFS - ARE WE THERE YET?
http://www.panasas.com/news/webinars/pnfs-are-we-there-yet

YouTube
"WEBINAR: pNFS-Are We There Yet?"
Panasas pNFS Update
2011/08/24, PanasasHPC, 50:35, 720p
http://www.youtube.com/watch?v=lRTu6VxRrQM

"NFS has been the standard protocol for NAS systems since the 1980s.
However, with the explosive growth of Linux clusters running
demanding technical computing applications, NFS is no longer
sufficient for these big data workloads. After years of development
effort, driven by Panasas and others, pNFS is now just around
the corner and promises to dramatically improve Linux client
I/O performance thanks to its parallel architecture.

Geoffrey Noer, Senior Director of Product Marketing at Panasas
presents this on-demand webinar -- "pNFS: Are We There Yet?""

Agenda:
*) Panasas Company Update
*) Evolution of NAS Architectures
*) The History of NFS up to NFS v4.1
*) pNFS Client and Server Architectures
*) Industry Support for pNFS
*) Status of pNFS Standard & Linux Implementation
*) Why Panasas for pNFS Storage
*) 最後の方は質疑応答

News & Events - Webinars
http://www.panasas.com/news/webinars
PanasasHPCチャンネル
http://www.youtube.com/user/PanasasHPC

参考ビデオ
"The Origin of Panasas"
2011/07/18, PanasasHPC, 2:38, 720p
http://www.youtube.com/watch?v=0DIOn7HA14w

"Panasas co-founder and CTO, Dr. Garth Gibson, talks about the early
days at Panasas and how the company grew to be the premier provider
of high performance parallel storage for technical computing
applications and big data workloads. Garth discusses the PanFS
parallel file system, Panasas scale-out NAS appliances, and
the future for competing file systems."

CEO Message
http://www.panasas.com/resource-library/videos/ceo-message
"Panasas Parallel Storage Vision"
2011/07/18, PanasasHPC, 2:13, 720p
http://www.youtube.com/watch?v=sM0QQoSkEtw

"Panasas CEO, Faye Pairman, discusses her industry background
the future of the company to an audience of HPC executives.
Faye talks about HPC market growth and the unprecedented demand for
scalable storage. She underscores the company's commitment
high performance computing and details how Panasas is uniquely
positioned to lead with scale-out storage appliances and the PanFS
parallel file system."
==================================================
[san-tech][03191] Panasas新製品：ActiveStor 11, Scalable performance and
capacity for cost-conscious HPC deployments
Date: Tue, 21 Jun 2011 00:27:40 +0900
[san-tech][02814] PAS12 HPCwireの解説記事＋Convey Computerとの

パートナー (Re: Panasas最上位機種 (PAS 12) 発表 (2010/11/10))
Date: Thu, 18 Nov 2010 12:51:58 +0900

the human genome, Nature, Vol. 409, No. 6822, 15 February 2001

2011-08-26T19:47:00.000+09:00

"Artificial Life", The Proceedings, September, 1987, etc.

2011-08-26T18:29:00.001+09:00

[san-tech][03356] 白色LEDで 100Mbpsデータ転送 (100Mbps by white-light LEDs)

2011-08-25T14:45:00.000+09:00

Date: Thu, 25 Aug 2011 14:17:15 +0900
--------------------------------------------------

EUのプロジェクトで白色 LED可視光を利用して 100Mbpsのデータ転送に

成功しました：

"Data are traveling by light"

Research News August 2011

Fraunhofer Institute for Digital Media Technology IDMT

http://www.fraunhofer.de/en/press/research-news/2010-2011/20/data-traveling.jsp

"Regular LEDs can be turned into optical WLAN with only a few

additional components thanks to visible light communication

(in short, VLC). The lights are then not just lighting up, they

also transfer data. They send films in HD quality to your iPhone

or laptop, with no loss in quality, quickly and safely."

.....

"... since scientists from the Fraunhofer Institute for

Telecommunications, Heinrich Hertz Institute HHI in Berlin,

Germany, have developed a new transfer technology for video data

within the scope of the OMEGA project of the EU, its

implementation in real life is getting markedly closer. At the

end of May, the scientists were able to present the results of

the project in Rennes, France. They were able to transfer data

at a rate of 100 megabits per second (Mbit/s) without any losses,

using LEDs in the ceiling that light up more than ten square

meters (90 square feet). The receiver can be placed anywhere

within this radius, which is currently the maximum range.

"In the future data will be transferred to laptops with the help of LEDs"

http://www.fraunhofer.de/en/Images/md08_fo2g_datentransfer_tcm63-94452.jpg

""For VLC the sources of light - in this case, white-light LEDs -

provide lighting for the room at the same time they transfer

information. With the aid of a special component, the modulator,

we turn the LEDs off and on in very rapid succession and transfer

the information as ones and zeros. The modulation of the light is

imperceptible to the human eye. A simple photo diode on the laptop

acts as a receiver.""

Klaus-Dieter Langer, the team of project manager, HHI

.....

"The new transmission technology is suitable for hospitals,

for example, because radio transmissions are not allowed there."

.....

"Currently the scientists are developing their systems toward

higher bit rates. "Using red-blue-green-white light LEDs, we were

able to transmit 800 Mbit/s in the lab," said Klaus-Dieter Langer."

OMEGA Project

http://www.ict-omega.eu/

[san-tech][03354] Gluster Announces Apache Hadoop Storage Compatibility, August 23rd, 2011

2011-08-23T23:02:00.000+09:00

Date: Tue, 23 Aug 2011 22:54:10 +0900
--------------------------------------------------

まだ GlusterFS 3.3 beta 2レベルですが、

"Gluster Announces Apache Hadoop Storage Compatibility in Latest

GlusterFS Release"

August 23, 2011

http://www.gluster.com/2011/08/23/gluster-announces-apache-hadoop-storage-compatibility-in-latest-glusterfs-release/

"... today announced the release of GlusterFS 3.3 beta 2 which

includes compatibility for Apache Hadoop. This latest enhancement

to GlusterFS uses the standard file system APIs available in Hadoop

to provide a new storage option for Hadoop deployments. With this

new feature, GlusterFS is the first open source, POSIX-compliant

file and object storage solution that works with Hadoop. Compatible

with existing MapReduce-based applications and with the ability

to co-exist with HDFS, this new functionality opens up data within

Hadoop deployments to any file- or object-based application,

providing a broad new range of options for big data deployments."

Features and benefits include (一部) :

*)"Fast access to data - with no centralized metadata or name server,

GlusterFS delivers better performance and easier manageability,

with no centralized point of failure;

*)"N-way replication - synchronous and asynchronous replication enables

users to replicate data, changes are tracked and queued to ensure

data stays synchronized regardless of latency or potential network

interruptions;

*)"Increased flexibility in sizing - with Gluster, Hadoop deployments

are no longer limited to large block sizes or 16 TB volume sizes;"

GlusterFS 3.3 Beta Resources

http://www.gluster.com/community/documentation/index.php/3.3beta

Hadoop - GlusterFS Resource Page

http://www.gluster.com/community/documentation/index.php/Hadoop

"Gluster Adds Hadoop Support to Storage Offering"

AUGUST 23RD 2011, SiliconANGLE

http://siliconangle.com/blog/2011/08/23/gluster-adds-hadoop-support-to-storage-offering/

==================================================

[san-tech][03307] GlusterFS OpenStack対応へ, July 27, 2011
Date: Thu, 28 Jul 2011 22:38:37 +0900

[san-tech][03202] Brightcove (Online Video Platform) が GlusterFSを採用
Date: Thu, 23 Jun 2011 15:24:22 +0900

[san-tech][03352] Slide: Titan Summit (Cray XK6, 20PF Peek), Oak Ridge Leadership Computing Facility (OLCF), (2011/08/15 - 17)

2011-08-21T12:41:00.002+09:00

Date: Sun, 21 Aug 2011 12:27:50 +0900
--------------------------------------------------
こちらも US DOE HPC関連ですが、Oak Ridge Leadership Computing Facilityの
次期 HPC: Titan (20PF Peak, Cray XK6, Opteron + GPU)についてのワークショップ
資料が公開されました：

Titan Summit
Oak Ridge Leadership Computing Facility (OLCF), August 15 - 17, 2011
http://www.olcf.ornl.gov/event/titan-summit/

"The summit will provide a forum for current and prospective users
to meet with OLCF staff and industry vendors to discuss Titan,
the next-generation leadership-class computational resource, and
a significant step towards exascale computing at the OLCF. Users
are encouraged to present their application plans for scaling
to exascale."

Presentationsボタンをクリックすると公開資料ページになります。

"Titan Overview"
Robert Whitten
Oak Ridge National Laboratory
http://www.olcf.ornl.gov/wp-content/uploads/2011/08/TitanSummit2011_Whitten.pdf
※Titanの概要です (19 Page)。

"Titan, path-forward to Multi-petaflop and Exaflop Computing -
Using Cray's Hybrid Multi-core Programming environment for porting
applications to Titan"
John Levesque, Cray CTO Office
Director of Cray's Supercomputing Center of Excellence
http://www.olcf.ornl.gov/wp-content/uploads/2011/08/TitanSummit2011_Levesque.pdf
※42MB、H/Wだけでなく、プログラミング環境・現状でのベンチマーク等々
盛りだくさんです。むしろ後者に重点を置いてます (137 Page)。

各種アプリケーションのポーティング例、GPUプログラミング等々

Cray XK6
http://www.cray.com/Products/XK6/KX6.aspx
http://www.cray.com/Assets/PDF/products/xk/CrayXK6Brochure.pdf

"Cray Unveils the Cray XK6 Supercomputer
New System Leverages the Latest Technologies From Cray, AMD and NVIDIA"
May 24, 2011
http://investors.cray.com/phoenix.zhtml?c=98390&p=irol-newsArticle&ID=1567269&highlight=

以前紹介しましたが、良くまとまっているので再度紹介：

"Cray XK6 super mates Opterons with Nvidia GPU workhorses"
24th May 2011, The Register
http://www.theregister.co.uk/2011/05/24/cray_xk6_gpu_supercomputer/
Cray's XE6 and XK6 and their Gemini interconnect
http://regmedia.co.uk/2011/05/22/cray_xe6_xk6_schematic.jpg
Cray's muscular XE6 blade
http://regmedia.co.uk/2010/05/24/cray_xe6_blade_server.jpg
==================================================
[san-tech][03031] 講演資料：Cray XE6 Workshop, NERSC (2011/02/7 -8)
Date: Mon, 11 Apr 2011 09:03:57 +0900

[san-tech][03351] Slide: US DOE Advanced CompuTational Software (ACTS) Collection (2011/08/16 - 19)

2011-08-21T12:34:00.000+09:00

Date: Sun, 21 Aug 2011 11:04:23 +0900
--------------------------------------------------

US DOEの HPC関連ワークショップの

12th Workshop on the DOE

Advanced CompuTational Software (ACTS) Collection

August 16-19, 2011

http://acts.nersc.gov/events/Workshop2011/

講演資料が公開されました。

はじめに Advanced CompuTational Software (ACTS) の紹介から

The DOE ACTS Collection

http://acts.nersc.gov/

"The DOE ACTS (Advanced CompuTational Software) Collection is

a set of DOE-developed software tools," ..... " that make it

easier for programmers to write high performance scientific

applications for high-end computers."

ACTS Tools

http://acts.nersc.gov/tools.html

で、冒頭に紹介した Webページから

About the Twelfth DOE ACTS Collection Workshop

"The four-day workshop will present an introduction to the ACTS

Collection for application scientists whose research demand

includes either large amounts of computation, complex software

integration, the use of robust numerical algorithms, run in

a large number of cores or combinations of these. The workshop

will include a range of tutorials on ACTS tools and tools

developed by the DOE SciDAC Program, discussion sessions aimed

to solve specific computational problems, and hands-on practices

using NERSC's high performance computers."

資料は以下の Webページで公開されています：

Twelfth DOE ACTS Collection Workshop Agenda

Scalable and Robust Computational Tools for High-End Computing

http://acts.nersc.gov/events/Workshop2011/Agenda.htm

概要：

"Introduction"

Tony Drummond

Computational Research Division

Lawrence Berkeley National Laboratory

http://acts.nersc.gov/events/Workshop2011/Talks/Introduction.pdf

"TAU Performance System"

Sameer Shende (TAU team, University of Oregon), et al.

"Overview of the Global Arrays

Parallel Software Development Toolkit"

Bruce Palmer (Pacific Northwest National Laboratory), et al.

とか。

スーパーマイクロプロセッサ、bit 臨時増刊、1991年 3月

2011-08-20T09:54:00.001+09:00

[san-tech][03350] Slide: US DOE 2011 Workshop on Architectures II: Exascale and Beyond: Configuring, Reasoning, Scaling (2011/08/08 - 10)

2011-08-19T09:17:00.004+09:00

Date: Fri, 19 Aug 2011 08:50:38 +0900
--------------------------------------------------
[san-tech][03349] Slide: US DOE 2011 Workshop on Architectures I:
Exascale and Beyond: Gaps in Research, Gaps in our Thinking (2011/08/02, 03)
Date: Fri, 19 Aug 2011 08:40:41 +0900
--------------------------------------------------
先に [03349] 紹介した

2011 Workshop on Architectures I: Exascale and Beyond: Gaps in Research,
Gaps in our Thinking
http://www.orau.gov/archI2011/default.htm

の関連ワークショップの資料が公開されています：

2011 Workshop on Architectures II: Exascale and Beyond: Configuring,
Reasoning, Scaling, 8/8/11 - 8/10/11
http://www.orau.gov/archII2011/default.htm

"The goal for this workshop will identify ways that we can abstract

architectures, in general, to meet the ever-changing complexities
inherent in exascale (and beyond). This workshop will be essential
in molding our path forward in hardware and software architectures
design space exploration, including simulation and emulation modeling
and their role in developing and implementing abstract machine models.
Moreover, we believe that such methodologies will prove insightful
in discovering what could and should be included or changed in next
and future generation hardware. Scholars from both DOE and non-DOE
communities will participate. Together we will identify machine
abstractions, their idealized primitives, and view of memory. It will
be informative to discuss and debate the realization of these
abstractions. Moreover, given an existing real system and/or subsystem,
how do we do explain its machine abstraction and would this abstraction
be sufficient to explain the performance attributes of exascale?"

Workshop on Architectures Iとはかなり趣きが異なります。
キーワードは "abstract"

Agenda
http://www.orau.gov/archII2011/agenda.htm

Presentations：
http://www.orau.gov/archII2011/presentations.htm

例えば (講演のほとんど、19編公開)

"Verification Challenges of Special-Purpose Hardware"
Jeffrey Kuskin
D. E. Shaw Research

"Machine Abstractions: Challenges for Future Execution Models"
Richard A. Lethin
Reservoir Labs, Inc.

"The Growing Mismatch between Hardware and the Underlying Abstract
Machine Model"
John Shalf
NERSC Advanced Technology Group, Lawrence Berkeley National Laboratory

[san-tech][03349] Slide: US DOE 2011 Workshop on Architectures I: Exascale and Beyond: Gaps in Research, Gaps in our Thinking (2011/08/02, 03)

2011-08-19T09:12:00.003+09:00

Date: Fri, 19 Aug 2011 08:40:41 +0900
--------------------------------------------------
以下のワークショップの講演資料が公開されています：

2011 Workshop on Architectures I: Exascale and Beyond: Gaps in Research,
Gaps in our Thinking
http://www.orau.gov/archI2011/default.htm

"It has been approximately two years since DOE and DARPA completed their
"ground breaking" exascale reports. In July 2010, the DARPA UHPC program
was started. Four teams were funded to research and develop ExtremeScale
designs. Various computer vendors have claimed that they are investigating
the designs of exascale computers (IBM, HP and SGI). Several DOE labs are
also investigating exascale designs (LBL, Sandia). The purpose of
this workshop is to address the hardware and software architecture
challenges for developing exascale computers. It would be very informative
to revisit the exascale studies and explore what has been learned and
to discuss research gaps. To achieve this goal we will meet and discuss
the lessons learned from these various activities."

Agenda
http://www.orau.gov/archI2011/agenda.htm

Presentations:
http://www.orau.gov/archI2011/presentations.htm

例えば (全 15講演、全て公開)、

"Silicon Photonics for Exascale Computing"
Keren Bergman
Lightwave Research Laboratory, Columbia University

"Research Gaps in Photonics, Memristive Memories, and Architecture"
Norman P. Jouppi
Senior Fellow and Director, HP Intelligent Infrastructure Lab

"Power and Programmability The Challenges of ExaScale Computing"
Bill Dally
Chief Schientist, NVIDIA
Bell Professional of Engineering, Stanford University

"Memory for Exascale and ... Micron's new memory component is called HMC:
Hybrid Memory Cube"
Dave Resnick
Sandia National Labs
  [san-tech][03160] "Open-Silicon and Micron Align to Deliver Next-Generation
  Memory Technology", June 7, 2011
Date: Wed, 08 Jun 2011 19:32:01 +0900

"Update on Current Trends and Roadblocks"
Peter M. Kogge
McCourtney Chair in Computer Science & Engineering
Univ. of Notre Dame
  [san-tech][03197] Graph 500ベンチマークをソフトウェアで高速化,
  E.T. International, Inc. (ETI)
Date: Wed, 22 Jun 2011 08:26:47 +0900
  [san-tech][03199] Re: Graph 500ベンチマークをソフトウェアで高速化,
  E.T. International, Inc. (ETI)
Date: Thu, 23 Jun 2011 00:28:56 +0900

関連記事：
"Taking a Disruptive Approach to Exascale"
August 18, 2011, HPCwire
http://www.hpcwire.com/hpcwire/2011-08-18/taking_a_disruptive_approach_to_exascale.html

[san-tech][02527] Re: US DOE DARPA 新 HPCプロジェクト (OHPC) 公募
Date: Tue, 10 Aug 2010 20:23:38 +0900
[san-tech][02406] US DOE DARPA 新 HPCプロジェクト (OHPC) 公募
Date: Wed, 23 Jun 2010 23:18:10 +0900
==================================================
[san-tech][03350] Slide: US DOE 2011 Workshop on Architectures II:

Exascale and Beyond: Configuring, Reasoning, Scaling (2011/08/08 - 10)
Date: Fri, 19 Aug 2011 08:50:38 +0900

[san-tech][03346] Slide: 2011 Gordon Summer Institute (2011/08/08 - 12)

2011-08-18T19:16:00.003+09:00

Date: Thu, 18 Aug 2011 12:32:45 +0900

--------------------------------------------------

First Flash Memory-based Supercomputerをうたい文句とした、SDSC Gordonの

Get Ready for Gordon - Summer Institute, August 8 - 11, 2011

http://www.sdsc.edu/Events/summerinstitute2011/index.html

2011 Gordon Summer Institute Agenda

http://gordon.sdsc.edu/_files/documents/gsi_2011_agenda.pdf

の講演資料が公開されました。以下の Webページからダウンロード可能です：

2011 Gordon Summer Institute

http://gordon.sdsc.edu/news_events/events/gordonsummerinstitute2011.html

Gordonについて説明すると話がさらに長くなるので、現行 H/Wについては

次の資料を参考にして下さい (と良いながら引用してますね)：

"Overview of the Gordon Architecture"

Shawn Strande, Gordon Project Manager SDSC (Joint GSI/CSIG session)

http://education.sdsc.edu/gordon2011/gsi/GordonArchitecture-GSI2011.pdf

Gordon Design Innovations (Page 3) から一部抜粋。[他ページ等から追加]

*)Intel Sandy Bridge Processor

*)Flash memory:

300 TB of high performance Intel flash via 64 I/O nodes made available

as a local resource to applications [4.8TB SSD per I/O node]

*)vSMP Memory Aggregation Software {32 nodes, 8 TF, 2 TB, 2 I/O Node]

*)3D torus interconnect [Mellanox]

*)Data Oasis: 100GB/s, 4 PB Lustre file system

Gordon I/O Node (Page 12 & 13)

Compute and I/O node Configuration (Page 14 - 17)

Networking (Page 22 - 31)

Gordon System Specification (Page 32)

Data Oasis (Page 33)

Gordon: Data-Intensive Supercomputing

http://gordon.sdsc.edu/

プレスリリース：

"SDSC Readying 'Gordon' Supercomputer for Pre-Production Trials this Month

First Flash Memory-based Supercomputer to Enter Production in January 2012

08.09.11

http://www.sdsc.edu/News%20Items/PR080911_gordon.html

※Gordonの位置づけの説明簡単にあるかもしれません。

NSFの Tera Gridの後継プロジェクト：XSEDE絡みの話も出てるのでやや

こしいかもしれません。

※つまり、私はちゃんと目を通してません・・・

開発計画初期については

[san-tech][01940] Re: SDSC: Triton Resource (Re: SDSC が Brocade HBA を採用)

2009/11/06

> "NSF Awards $20 Million to SDSC to Develop "Gordon""

> 11.04.09

> http://www.sdsc.edu/News%20Items/PR110409_gordon.html

[san-tech][02088] Re: SDSC: Triton Resource (Re: SDSC が Brocade HBA を採用)

2010/02/15

> "Getting Intense About Data: The SDSC Gordon Cluster"

> Douglas Eadline, Ph.D., December 15th, 2009

> http://www.linux-mag.com/id/7645/

とか

[san-tech][03345] Xsigo Server Fabric, August 15, 2011

2011-08-18T19:07:00.006+09:00

Date: Wed, 17 Aug 2011 20:58:59 +0900

--------------------------------------------------

Xsigoプレスリリースやビデオの前に、後半で紹介する The Registerの記事を

先に読んだ方が解りやすい気もしますが、まずはプレスリリースから

"Xsigo First to Market with Virtualized Server Fabric to Enable Cloud

Data Centers"

Xsigo Server Fabric lets enterprise IT managers and cloud providers

provision new services 10x faster at half the cost of operation

Xsigo leapfrogs switch and port-based legacy vendors Cisco, Brocade

and Juniper with fully virtualized network connectivity

August 15, 2011

http://www.xsigo.com/news_events/press_releases.php?pr=68

"Xsigo Systems today announced the Xsigo Server Fabric, the industry's

first fully virtualized infrastructure for cloud-optimized data centers.

The Xsigo Server Fabric enables one-click network connections from

virtual machines to any data center resource - including servers,

networks, storage, and other virtual machines - to accelerate

the management of virtualized data centers."

.....

"The Xsigo Server Fabric lets customers flexibly deploy Ethernet and

Fibre Channel connections at speeds up to 40 Gbps, resulting

in significant application acceleration. Environments can be scaled

to 1,000 physical hosts with tens of thousands of virtual connections

linking virtual machines to each other and to network and storage

resources. Because the Server Fabric is fully interoperable with

existing core networking products from Cisco, Brocade, Juniper

and others, and because it enables greater utilization of those assets,

customers can extend the life of their existing hardware to minimize

disruption and reduce capital costs."

"Xsigo Server Fabric"

2011/08/14, XsigoVirtualIO, 5:05, 360p

http://www.youtube.com/watch?v=93pUiuZH9yY

"Up to 30X Performance Gain, No Networking Pain"

Imagine 10X to 30X more performance in your I/O intensive apps,

19X faster vMotion, 12X faster DB queries, and backups done in minutes,

not hours.

August 14th, 2011, Xsigo Blog

http://www.xsigo.com/blog/?p=2240

Xsigo Server Fabric

http://www.xsigo.com/products/server_fabric.php

http://www.xsigo.com/_downloads/Brochures/server_fabric_datasheet.pdf

Without Xsigo and With Xsigo

http://www.xsigo.com/_img/server_fabric_before_after.png

Black & Red [02]

2011-08-17T00:05:00.000+09:00

[san-tech][03344] SmartOS from Joyent (ZFS, DTrace, KVM (OS-based virtualization (Zones)), Crossbow)

2011-08-16T22:57:00.000+09:00

Date: Tue, 16 Aug 2011 22:35:03 +0900
--------------------------------------------------

Sun DTraceの開発功績で、Technology Review, MITの TR35に選出された

Bryan Cantrill氏が移籍した Joyentから、(illumosベースの) SmartOSが

一般公開されました：

"Joyent Announces SmartOS With KVM: An Open Source, Modern Operating System"

KVM on SmartOS provides entrepreneurs and enterprises highest performance

and profit potential through best of breed cloud technologies

August 15, 2011

http://www.joyent.com/2011/08/joyent-announces-smartos-with-kvm-an-open-source-modern-operating-system/

"... today announced that it has ported KVM to its operating system,

SmartOS. SmartOS turns any server into a highly efficient,

multi-tenant application hosting platform. The open source project

offers innovative tools for application developers looking for

reliable, resilient storage and efficient virtualization that

Joyent has tuned and hardened for modern, data-intensive real-time

(DIRTy) application deployment."

※このプレスリリースでは Illumosについて触れてません。

概念図：

SmartOS: The Complete Modern Operating System

http://smartos.org/

"SmartOS incorporates the four most revolutionary OS technologies of

the past decade - Zones, ZFS, DTrace and KVM - into a single operating

system,"

OSes Booted under SmartOS Native KVM…to date

SmartOS 64-bit, Linux 64-bit, Windows Server 2008 R2, FreeBSD 8.1 32-bit

Plan 9 from Bell Labs, Seventh Edition, FreeBSD 4.9, Haiku OS, OpenBSD

QNX 4.25 (from a 1.44 MB floppy image), Weenix

このページに

"SmartOS is comprised of the Illumos kernel (with ZFS, DTrace, OS-level

virtualization and next-generation KVM) with BSD package management and

a GNU toolchain."

当然ながら

Joyent SmartOS FAQ

http://www.joyent.com/products/smartos/smartos-faq/

What kind of OS license is SmartOS using?

"SmartOS is distributed under the file-based CDDL license, and under

the MIT license where stated. The KVM work is licensed under GPL."

前置き (？) が長くて申し訳ありませんが以下のスライドがお勧め：

(itoooonさん、有り難うございます)

"Experiences porting KVM to SmartOS"

Bryan Cantrill, Vp, Engineering, Joyent

KVM Forum 2011, August 15th 2011

http://www.slideshare.net/bcantrill/experiences-porting-kvm-to-smartos

※登録ユーザはダウンロード可能

以下はスライドから

WTF is SmartOS? (Page 1)

As an illumos derivative, has several key features:

ZFS:

DTrace:

OS-based virtualization (Zones):

Network virtualization (Crossbow):

Enhancement: OS Virtualization (Page 22)

*)illumos has deep support for OS virtualization

*)While our implementation does not require it, we run KVM guests in

a local zone, with the QEMU process as the only process

*)This was originally for reasons of accounting

*)...but given the recent KVM vulnerabilities, it has become a matter

of security

Enhancement: Network virtualization (Page 23)

*)We create a virtual NIC (VNIC) per KVM guest

Enhancement: DTrace, cont. (Page 27)

*)Orthogonal to this work, we have developed a real-time analytics

framework that instruments the cloud using DTrace and visualizes

the result

※itoooonさん、判明したらフォローお願いします。

　これだけ DTraceを前面に出しているのだから、標準と期待しています。

[san-tech][03299] 2011/08/25(木)：SS研 HPCフォーラム2011　「ペタスケール HPC技術が支える地球科学」

2011-08-15T17:00:00.003+09:00

Date: Mon, 25 Jul 2011 12:54:55 +0900
--------------------------------------------------
2011年 8月 25日(木)、サイエンテフィック・システム研究会 (SS研) により、
SS研 HPCフォーラム 2011
ペタスケールHPC技術が支える地球科学
http://www.ssken.gr.jp/MAINSITE/activity/sectionmeeting/sci/2011-1/program.html
http://www.ssken.gr.jp/MAINSITE/activity/sectionmeeting/sci/2011-1/program.pdf
が開催されます。
※申込み：Webページ経由　 [受付期間:7/22(金)～8/15(月)]

日　　時：2011年8月25日(木) 分科会 10:30-17:25(受付9:45-)/懇親会 17:45-19:00
場　　所：汐留シティセンター 24階富士通株式会社大会議室
参加対象：SS研会員に限らずどなたでもご参加いただけるオープンイベント
定　　員：200名
参加費　：分科会:無料 / 懇親会:\500 (当日お支払い)
その他：室温は28℃設定です。服装はクールビズをお勧めします。
申込み：Webページ経由　 [受付期間:7/22(金)～8/15(月)]

講演題目
[講演1:津波] 基調講演「東日本大震災での津波の被害像と今後の減災への課題」
[講演2:固体地球] 「地磁気シミュレーションとバーチャルリアリティ可視化」
[講演3:気象] 「エクサスケール時代の気候モデル」
[講演4:地震] 「構造物と都市の地震シミュレーション」
[講演5] 「次世代スパコン『京』について」

[san-tech][03342] 次世代スーパーコンピュータ「京」ビデオ、ディスカバリーチャンネル＆BS-TBS

2011-08-12T23:52:00.002+09:00

Date: Fri, 12 Aug 2011 23:40:37 +0900
--------------------------------------------------

少し前のビデオですが、富士通さんサイトで以下の 2本のビデオが公開されています：

動画で見る、「次世代スーパーコンピュータ」

http://jp.fujitsu.com/about/tech/k/movies/index.html

"日本発　次世代スーパーコンピュータ「京」"

ディスカバリーチャンネル (2011年 3月 21日初回放送)

5分 59秒、ダイジェスト版 (英語＋日本語字幕等)

"夢をカタチにするスーパーコンピュータ"

BS-TBS (2010年 12月 18日放送)

6分 23秒、ダイジェスト版 (日本語)

「次世代スーパーコンピュータ」

http://jp.fujitsu.com/about/tech/k/

--------------------------------------------------

関連：
HPCI計画推進委員会, 文部科学省
http://www.mext.go.jp/b_menu/shingi/chousa/shinkou/020/gaiyou/1296773.htm

"今後のHPC技術の研究開発を検討する作業部会の開催について"

平成23年8月4日, 文部科学省
http://www.mext.go.jp/b_menu/houdou/23/08/1309406.htm

"　文部科学省では、HPCI計画推進委員会のもとに「今後のハイパフォーマンス・

コンピューティング（HPC）技術の研究開発の検討ワーキンググループ」を設置し、

今後のHPC技術の研究のあり方について検討しました。その結果を踏まえ、今後の

具体的な取組みとして、文部科学省及び関係機関の主催によるアプリケーション

作業部会及びコンピュータアーキテクチャ・コンパイラ・システムソフトウェア

作業部会を開催し、今後追求すべきHPCシステム案とこれを開発していく体制案を

とりまとめていきます。"

"次世代スパコン開発へ　計算速度「京」の100倍"

文科省、省エネ技術を後押し

2011/8/7, 日本経済新聞

http://www.nikkei.com/news/headline/article/g=96958A9C93819595E2E1E2E3998DE2E4E2EAE0E2E3E39F9FEAE2E2E2

[san-tech][03341] Slide: Flash Memory Summit 2011 (2011/08/09 - 11)

2011-08-12T05:00:00.005+09:00

Date: Fri, 12 Aug 2011 04:36:15 +0900
--------------------------------------------------

2011年 8月 9日～ 11日に開催されていた：

Flash Memory Summit 2011 Proceedings, August 9 - 11, 2011

http://www.flashmemorysummit.com/

2011 Program At-a-Glance

http://www.flashmemorysummit.com/English/Conference/Program.html

Preview Program

http://www.flashmemorysummit.com/English/Conference/Preview_Program.html

http://www.flashmemorysummit.com/English/Collaterals/Documents/FMS11_Preview_Program.pdf

の Slideが公開されています。

以下のページにメールアドレスを登録することで、ダウンロードページ

(Flash Memory Summit 2011 Proceedings) にジャンプします。

Get the Flash Memory Summit Proceedings

http://www.flashmemorysummit.com/cgi-bin/start.cgi/HTMLOS_Pages/Entrance_Proceedings.html

現時点では一括ファイルは用意されてませんが (確か去年はあった気が)、

1セッションを除いた 100ファイルが公開されています。内容は多岐に渡って

いますし、空白ページがあったり、資料内容の深さも様々です。

登録者向けのページなので講演名等は控えますが、今まで投稿してある程度

関連しそうなもの (順不同、ただし、講演との関連はいろいろです)

[san-tech][03149] UCSD: Phase Change Memory-Based SSD
Date: Sat, 04 Jun 2011 01:40:13 +0900

[san-tech][03118] Slidecast: "IO Turbine Launches Accelio SSD Accelerator

for Virtualization", 05.24.2011, insideHPC,com

Date: Thu, 26 May 2011 14:17:11 +0900

[san-tech][03152] Re: Slidecast: "IO Turbine Launches Accelio SSD Accelerator

for Virtualization", 05.24.2011, insideHPC,com
Date: Sun, 05 Jun 2011 17:30:13 +0900

[san-tech][03317] SSDの信頼性 ("Investigation: Is Your SSD More Reliable

Than A Hard Drive?", July 29, 2011, Tom's Hardware, )
Date: Sun, 31 Jul 2011 23:43:46 +0900

[san-tech][03200] SolidFire 1U SSD: Real time compression/deduplication,

Scalability to 100 node
Date: Thu, 23 Jun 2011 12:49:33 +0900

[san-tech][02384] DOE LLNL Hyperion (Appro + Fusion-IO) &

Virident PCIe SSDボードとか
Date: Thu, 17 Jun 2010 08:54:37 +0900

[san-tech][02826] Virident tachIOn (PCIe SSD)：住友情報システムが代理店契約
Date: Wed, 24 Nov 2010 17:12:57 +0900

以下は Flash Memory Summitに直接関連する記事だけではありません：

※初めの 2つ：SiliconANGLE と Wikibon Blogは切り口として言及してます。

"FlashSoft - Storage is Sexy But Cache Is King"

AUGUST 10TH 2011, SiliconANGLE

http://siliconangle.com/blog/2011/08/10/flashsoft-storage-is-sexy-but-cache-is-king/

FlashSoft

http://flashsoft.com/

"FlashSoft viewpoint: The importance of the Fusion IO + IO Turbine acquisition"

日付無し、Blog

http://flashsoft.com/blog/fusion-io-io-turbine-acquisition.html

-----

"Fusion-io: A Flash Bird of a Different Feather"

August 10, 2011, Wikibon Blog (ツイートで情報を頂きました)

http://wikibon.org/blog/fusion-io-a-flash-bird-of-a-different-feather/

.....

"Here's why they're wrong. Fusion-io's advantage has little to do

with PCIe and everything to do with software, specifically its

Virtual Storage Layer (VSL) technology. This is why, in addition

to its hyper-growth and killer customers, the company in my mind

can justify a $2B valuation. Wikibon's David Floyer explains VSL

in gory detail in this excellent post."

"Fusion-io Revolutionizes Flash Storage Management with VSL"

Last Update: Aug 07, 2011, Originating Author: David Floyer

http://wikibon.org/wiki/v/Fusion-io_Revolutionizes_Flash_Storage_Management_with_VSL

[san-tech][03108] Fusion-io VSL (Virtual Storage Layer) 関連の国際会議での発表
Date: Tue, 24 May 2011 17:28:44 +0900

-----

"Nexsan plays catchup with flash-cached storage controller"

E5000 with added flash stash

11th August 2011, The Register

http://www.theregister.co.uk/2011/08/11/nexsan_flashy_e5000/

"Nexsan has introduced an E5000 NAS storage array family with two tiers

of solid state controller storage to boost performance, following in

the footsteps of NetApp, with its Flash Cache, and Oracle."

"Nexsan Introduces E5000 Family Of NAS Storage Systems With FASTier

For Accelerated Performance"

E5000 Family leverages multiple tiers of SSD technology to cache file

system data, resulting in a 6X performance increase for 8 TB to 720 TB

of capacity with industry-leading storage density and power management

August 9, 2011

http://www.nexsan.com/about/news/pressreleases/2011/080911.aspx

Nexsan E5000

http://www.nexsan.com/products/e5000/index.aspx

h ttp://www.nexsan.com/products/library/Nexsan_E5000_DS.pdf

http://www.nexsan.com/products/library/Nxsn_E5000_StorageSystems_SpecSheet.pdf

-----

"Samsung and STEC flash their sizzling SSDs"

Storage action remains flashy

11th August 2011, The Register

http://www.theregister.co.uk/2011/08/11/stec_and_samsung_ssds/

-----

"SMART unveils smarter, faster, fatter SSDs"

'How much for the 1.6TB baby? We're not saying'

4th August 2011, The Register

http://www.theregister.co.uk/2011/08/04/smart_optimus_ssd/

Date: Tue, 16 Aug 2011 03:12:42 +0900

--------------------------------------------------
Flash Memory Summit 2011について Wikibonでの David Floyerさんの記事です：

"2011 Flash Memory Summit Enterprise Roundup"
Aug 15, 2011, Wikibon
Originating Author: David Floyer
http://wikibon.org/wiki/v/2011_Flash_Memory_Summit_Enterprise_Roundup

Contents
1 In Process - please wait until this line disappears to edit/tweet!
1.1 Sentiment at the Flash Summit
1.2 The Four Horsemen
1.2.1 Traditional SSD in an Array
1.2.2 Flash Cache
1.2.3 Flash-only Arrays
1.2.4 Flash as an Extension of RAM

すいません、中身は見てません・・・

[san-tech][02826] Virident tachIOn (PCIe SSD)：住友情報システムが代理店契約

2011-08-12T04:57:00.001+09:00

Date: Wed, 24 Nov 2010 17:12:57 +0900
--------------------------------------------------
住友情報システムさんが、Fusion-ioとは別の PCIeベース SSD専門会社の
Virident社と代理店契約を結びました：

"Virident Systems Expands Global Reach to Japan with Sumisho Computer
Systems", Nov. 15, 2010
http://www.virident.com/news_10_11_15_sumisho.php

"announced a new partnership with Sumisho Computer Systems Corporation
(SCS), a leading provider of solutions and services to the Japanese
enterprise and high-performance computing (HPC) market. In partnering
with SCS, Virident will leverage the Tokyo-based firm's expertise in
the Japanese enterprise market and gain a local provider of
enterprise-class systems."

tachIOn Solid State Disk
http://www.virident.com/products.php
http://www.virident.com/downloads/Virident_tachIOn_Specifications.pdf
http://www.virident.com/downloads/Virident_tachIOn_Drive_Product_Brief.pdf

tachIOn Key Features (一部):
*)Enterprise-grade reliability: Advanced ECC; Dynamic global wear leveling
*)Field Serviceability, upgrade path: Replaceable flash modules
*)Industry-leading lifetimes: 24 years at 5TB writes/day; Extend investment
with replaceable modules
*)Support for leading NAND Flash vendors: Samsung, Micron, Toshiba

住友情報システム
http://www.scs.co.jp/
※Virident代理店についてのプレスリリースは出ていませんね。

他のプレスリリース
"Virident Systems Announces 800 GB tachIOn SSD", Nov. 15, 2010
http://www.virident.com/news_10_11_15_ssd.php

"today announced that the tachIOn・Solid State Disk (SSD) is now
available with 800 GBs of usable capacity."
...
"NERSC purchased 20 of the 400GB cards, a total of 8 TB, and is in the
process of installing them and making them available to the center's
3,000-plus researchers."

2011/09/10 (土) 「三好甫先生記念計算科学シンポジウム」

2011-08-10T18:00:00.002+09:00

～わが国スーパーコンピュータの来し方・未来を語る～一日として
「三好甫先生記念計算科学シンポジウム」を開催いたします．
http://www.hpfpc.org/miyoshi-sympo/

講演会・懇親会　参加登録
http://www.hpfpc.org/miyoshi-sympo/touroku.html

参加申込方法：
事前登録へのご協力をお願いいたします．
8月10日(水)までにお申し込みください。

==================================================

[san-tech][02964] 「地球シミュレータ開発史」公開 (2011/02/17)
Date: Fri, 18 Feb 2011 18:22:35 +0900

Take a Bath!

2011-08-10T03:20:00.000+09:00

Laleham Park, England, 2009/11

2011-08-09T21:59:00.000+09:00

[san-tech][03339] SIGGRAPH 2011でも Fusion-io, Aug. 8, 2011

2011-08-09T20:10:00.002+09:00

Date: Tue, 09 Aug 2011 20:05:41 +0900
--------------------------------------------------

Fusion-ioのプレスリリースです：

"Fusion-io Teams Up with NVIDIA, Tweak Software and Thinkbox Software

to Accelerate Entertainment Production"

Innovative Industry Leaders Unlock Performance to Unleash Creativity

in Digital Content Creation Tools Showcased at SIGGRAPH 2011

Aug. 8, 2011

http://www.fusionio.com/press-releases/fusion-io-teams-up-with-nvidia-tweak-software-and-thinkbox-software-to-accelerate-entertainment-production/

http://www.fusionio.com/load/-media-/1d7gzn/docsPress/FINALPR_FIO_Siggraph_8.8.11.pdf

"... today announced its participation in the SIGGRAPH 2011

international conference on computer graphics and interactive

technologies. Fusion-io is collaborating with a number of leading

technology companies, including NVIDIA, Thinkbox Software and

Tweak Software, to accelerate entertainment production

by demonstrating full resolution, real-time digital content creation

for many of the industry's most powerful applications."

各ソフトウェアメーカーへのリンクはプレスリリース中にあります。

SIGGRAPH 2011

http://www.siggraph.org/s2011/

※CG系では世界最高級の国際会議です。

[san-tech][03337] IBM: NCSA Blue Watersキャンセルだ！, Aug. 8, 2011, Effective August 6, 2011

2011-08-09T03:42:00.008+09:00

Date: Tue, 09 Aug 2011 03:39:33 +0900
--------------------------------------------------

※Blue Waters Interconnect Network情報追加 (09 Aug 2011 21:13:59 +0900)

あれまぁ、

"Blue Waters Update-NCSA/IBM Joint Statement"

Aug. 8, 2011, Effective August 6, 2011

http://www.ncsa.illinois.edu/BlueWaters/system.html

"IBM terminated its contract with the University of Illinois to provide

the supercomputer for the National Center for Supercomputing Applications'

Blue Waters project."

"NCSA is confident that its goal of building a sustained-petascale

supercomputer remains achievable in a timely manner. NCSA is

coordinating with the National Science Foundation to ensure project

continuity and that the goals of the project are achieved."

.....

"IBM will return money received to date and NCSA will return equipment

delivered by IBM per terms of the contract."

関連記事 (と言っても上のプレスリリースがメイン)

"IBM Cancels Blue Waters"

Aug 8, 2011, insideHPC

http://insidehpc.com/2011/08/08/ibm-cancels-blue-waters/

2011/08/09

"IBM yanks chain on 'Blue Waters' super"
Power7 petaflops behemoth gets flushed
8th August 2011, The Register
  http://www.theregister.co.uk/2011/08/08/ibm_kills_blue_waters_super/

"IBM Bails on Blue Waters Supercomputer"
August 08, 2011, HPCwire
  http://www.hpcwire.com/hpcwire/2011-08-08/ibm_bails_on_blue_waters_supercomputer.html

--------------------------------------------------
Date: Tue, 09 Aug 2011 21:13:59 +0900

今さらながら、NCSA Blue Waters Interconnect Network についてのおさらい：

"Blue Waters:
Leading the Way in Sustained Petascale Computing"
Thom Dunning, Director,
National Center for Supercomputing Applications (NCSA),
University of Illinois at Urbana-Champaign (UIUC), USA
DEISA PRACE Symposium (13-14 April, 2011, Helsinki, Finland)
  http://www.prace-project.eu/documents/dunning-pdf.pdf

*)Logical View of Blue Waters Interconnect (Page 9 - 11)

*)Two-level (L, D) Direct-connect Network (Page 12)

Each Supernode = 32 QCMs (4 Drawers x 8 SMPs/Drawer)
Fully Interconnected with Llocal and Lremote Links
*)Lremote links between drawers in supernode (240 GB/s)
&
Blue Waters = 320 Supernodes (40 BBs x 8 SNs/BB)
Fully Interconnected with D Links
*)D-links between supernodes (320 GB/s)
*)BB: 3-Rack Building Block (Each BB = 8 Supernodes)

*)Blue Waters vs. Other Future Systems (Page 15)
Network
Interconnect Topology Bisection Bandwidth (TB/s) [Peak PF]
NCSA Blue Waters (IBM PERSC, IBM Power 7): 2-Level Direct, ~500 [10]
ANL Mira (IBM BlueGene, PowerPC): 5D Torus, 30** [10]
RIKEN K-Computer (Fujitsu SPARC64 VIIIfx): Mesh/Torus >30 [>10]

Documents, Partnership for Advanced Computing in Europe (PRACE)
http://www.prace-project.eu/documents
==================================================
[san-tech][03220] IBM Power 775：光ファイバー (120Gbps) で筐体間を接続
Date: Tue, 28 Jun 2011 22:20:49 +0900

[san-tech][03336] New EMC Disk Library for Mainframe, The industry’s fastest mainframe VTL for z/OS, August 8, 2011

2011-08-09T03:23:00.000+09:00

Date: Tue, 09 Aug 2011 03:10:06 +0900
--------------------------------------------------

EMCのプレスリリースです：

"New EMC Disk Library for Mainframe is First Integrated Solution

for All Tape Use Cases"

EMC DLm6000 2x Faster than Competition and Reduces Reliance

on Mainframe Tape

August 8, 2011

http://www.emc.com/about/news/press/2011/20110808-01.htm

News Summary

*)The new DLm6000 is the only solution that can address the full

range of mainframe tape workloads with a single, consolidated

all-disk system

*)Designed for use in IBM z/OS environments, the new EMC DLm6000

is the industry's fastest mainframe VTL

*)Integrates EMC VNX7500 and/or EMC Data Domain DD890 storage

systems to support all mainframe use cases

*)Matches different workloads to the most appropriate storage

Disk Library for mainframe - Virtual Tape Library, Backup Storage - EMC

http://www.emc.com/backup-and-recovery/disk-library/disk-library-for-mainframe.htm

http://www.emc.com/collateral/hardware/data-sheet/h4207-emc-disk-library-mf-ds.pdf

"EMC trundles out deduping mainframe VTL"

Bus-Tech allied to Data Domain

8th August 2011, The Register

http://www.theregister.co.uk/2011/08/08/emc_bus_tech_mainfram_vtl/

"The DLM6000 hooks up to a Z/OS mainframe and can use back-end

VNX storage and/or a Data Domain 890. EMC says it's a single product,

the only single product in fact, that can cover "the full range of

mainframe tape workloads with a single, consolidated all-disk system."

It talks about tape workloads such as "backup and recovery,

batch processing, DFHSM migration and data archive", and says

mainframe users no longer have to have multiple systems doing

these tape-related things. That sounds pretty compelling."

"EMC's Stuck on Tape, Speeds Mainframe Performance"

AUGUST 8TH 2011, SiliconANGLE

http://siliconangle.com/blog/2011/08/08/emcs-stuck-on-tape-speeds-mainframe-performance/

"EMC's DLm6000 is aiming to be an all-in-one solution in solving

numerous challenges that mainframe users face in dealing with

physical tape systems."

==================================================
[san-tech][02806] "EMC Acquires Bus-Tech", November 10, 2010
Date: Thu, 11 Nov 2010 23:24:40 +0900

[san-tech][02806] "EMC Acquires Bus-Tech", November 10, 2010

2011-08-09T03:18:00.000+09:00

Date: Thu, 11 Nov 2010 23:24:40 +0900
--------------------------------------------------

"EMC Acquires Bus-Tech", November 10, 2010

http://www.emc.com/about/news/press/2010/20101110-02.htm

"... today announced it has acquired Bedford, Massachusetts-based

Bus-Tech, Inc. Bus-Tech is a privately held provider of VTL

(Virtual Tape Library) solutions that utilize open systems disk

storage to store and retrieve mainframe tape data. Bus-Tech

products enhance EMC solutions for mainframe batch processing,

backup and recovery, disaster recovery, and data archiving

applications. Bus-Tech now becomes part of the rapidly growing

EMC Backup Recovery Systems division, which delivers next-generation,

disk-based backup and recovery solutions. The acquisition is not

expected to have a material impact to EMC GAAP and non-GAAP EPS

for the full 2010 fiscal year."

.....

"Bus-Tech has been a member of the EMC Select partner program

since 2004, and has been an EMC OEM partner since March of 2008.

Bus-Tech most recently collaborated with EMC in July of 2010 to

help deliver the Deduplication Storage Expansion option for the

EMC DLm960 Disk Library for Mainframe, which was based on the

widely adopted EMC Data DomainR DD880 deduplication storage system.

The pairing of Bus-Tech mainframe"

EMC DLm960 Disk Library for Mainframe

http://www.emc.com/collateral/software/specification-sheet/h5937-emc-disk-lbry-mf-ss.pdf

EMC Data Domain DD880

http://www.datadomain.com/pdf/DataDomain-DD880-Datasheet.pdf

Bus-Tech,Inc

http://www.bustech.com/

"EMC Buys Mainframe VTL Maker Bus-Tech"

November 10, 2010, Enterprise Storage Forum.com

http://www.enterprisestorageforum.com/continuity/news/article.php/3912471

"EMC buys into mainframe VTL market"

11th November 2010, The Register

http://www.theregister.co.uk/2010/11/11/emc_buys_bus_tech/

London Eye, 2009/11

2011-08-07T22:17:00.001+09:00

[san-tech][03335] OpenDRAC (Re: "Ciena's Dynamic Resource Allocation Controller Released As Open Source Software", 2010/06/03)

2011-08-07T20:26:00.000+09:00

Date: Sun, 07 Aug 2011 20:18:37 +0900
--------------------------------------------------

一年以上前の投稿への自己レスですが、Ciena Dynamic Resource Allocation

Controller (DRAC)、元は Nortelの技術は、既に

OpenDRAC

the Open Dynamic Resource Allocation Controller

https://www.opendrac.org/

サイトが用意され、すでに OpenDRAC 4.1.0 (June 2011) が公開されてました。

Download

https://www.opendrac.org/?page_id=79

Documentation

https://www.opendrac.org/?page_id=57

Blog

https://www.opendrac.org/?page_id=178

Cienaの中の Blog

"Q&A with Ciena DRAC expert Paul Runnels"

19 Jul 2010, Ciena Insights

http://www.ciena.com/corporate/blog/QA-with-Ciena-DRAC-expert-Paul-Runnels.html

"Ciena DRAC Goes Open Source for Network Control"

Ciena open sources Dynamic Resource Allocation Controller,

technology it acquired from bankrupt Nortel technology

for managing network path control.

Jun 7, 2010, EnterpriseNetworkingPlanet

http://www.enterprisenetworkingplanet.com/netsysm/article.php/3886431/Ciena-DRAC-Goes-Open-Source-for-Network-Control.htm

.....

"Ciena is open sourcing a technology called the Dynamic Resource

Allocation Controller (DRAC) under the GPLv3 license. DRAC provides

a solution for abstracting network resources and enabling users

to provision dynamic on-demand lightpaths.

The approach could provide a new method for connecting large

and disparate networking assets."

.....

"Ciena acquired the DRAC technology through its $774 million purchase

of Nortel's optical networking and Metro Ethernet business units."

.....

"DRAC is already in use on Ciena's commercial equipment today and

Runnels said that he expects it to be a mainstay in its portfolio going

forward. Now that the DRAC technology is open source it could also

end up in products from other vendors in the future as well.

Ultimately, Ciena envisions DRAC growing into a user-provisioned

network services tool supported by multiple vendors."

Paul Runnels, a member of Ciena's product line management team

Research and education organizations限定ではなかったようですね。

On Wed, 16 Jun 2010 21:09:50 +0900

--------------------------------------------------

> Cienaからプレスリリースがありました：

> "Ciena's Dynamic Resource Allocation Controller Released As Open Source Software"

> 06/03/2010

> http://www.ciena.com/news/news_25150.htm

> ただし、

> "Research and education organizations can now accelerate development

> of dynamic and on-demand end user control of network resources for

> high performance network applications"

> "today announced that the company's Dynamic Resource Allocation

> Controller (DRAC), intelligent middleware for provisioning dynamic

> on-demand lightpaths across different networks, has been released as

> open source software. By publicly publishing the source code,

> individuals and organizations in the research and education community

> can collaboratively contribute and evolve the software to support the

> advanced networking requirements of scientific researchers worldwide."

> ....

> "For additional information on Ciena's collaboration with the research

> and education community, please visit:

> http://www.ciena.com/researchandeducation."

> で、

> Ciena - Network solutions for the research and education industry

> http://www.ciena.com/researchandeducation

> なんですが、Open Source関連の情報は見つかりません。

> Dynamic Resource Allocation Controller (旧称 Dynamic Bandwidth Allocation

> (DBA)？) 関連資料 (White Paper)

> Search results: Dynamic Resource Allocation Controller

> http://search.ciena.com/search?p=Q&lbc=ciena&uid=89608389&ts=custom&w=Dynamic%20Resource%20Allocation%20Controller&af=cat1:resources&method=and

> White Paperへのアクセスには登録が必要です。

[san-tech][03334] Slide/Video: NIST Cloud Computing Forum & Workshop III (2011/04/07, 08)

2011-08-07T19:27:00.001+09:00

Date: Sun, 07 Aug 2011 18:57:29 +0900
--------------------------------------------------

2011年 4月 7、8日に開催された

NIST Cloud Computing Forum & Workshop III

http://www.nist.gov/itl/cloud/cloudworkshopiii.cfm

Agenda

http://www.nist.gov/itl/cloud/upload/April_2011_CCWWIII_agenda.pdf

http://collaborate.nist.gov/twiki-cloud-computing/bin/view/CloudComputing/Workshop3Agenda

※下は Bioへのリンク付き

のハイライトビデオとスライドが公開されています。

ハイライトビデオ：

http://www.ebmcdn.net/niststream/flash/Cloud11/cloud11-playlist.html

スライド：

Useful Documents for Cloud Adopters, NIST

http://collaborate.nist.gov/twiki-cloud-computing/bin/view/CloudComputing/Documents

真ん中あたりに

Forum and Workshop Presentation Materials

Forum and Workshop III, April 2011

があります。

NIST Cloud Computing Program

http://www.nist.gov/itl/cloud/

Information Technology Laboratory, NIST

http://www.nist.gov/itl/

例えば：

ITL Quantum Information Program

http://www.nist.gov/itl/quantum/index.cfm

先に紹介した Forum資料のすぐ上に

NIST Special Publication and Reference Materials

があります。以前ドラフト版を紹介した以下の最終版が公開されています：

"Guide to Security for Full Virtualization Technologies"

Karen Scarfone, Murugiah Souppaya and Paul Hoffma

January 2011, NIST Special Publication 800-125

http://csrc.nist.gov/publications/nistpubs/800-125/SP800-125-final.pdf

[san-tech][02485] US NIST:Guide to Security for Full VirtualizationTechnologies

(Draft)、2010/07/26

> 日本語の解説 (翻訳)

> "NISTがGuide to Security for Full Virtualization Technologiesのドラフト

> を公開（20100722-3）"

> http://virtualization.info/jp/news/2010/07/nistがguide-to-security-for-full-virtualization-technologiesのドラフトを公開（20100722-3）.html

> "35ページ構成の同書は3つのセクションに分かれており、1つめのセクション

> はフル仮想化とそのインプリメンテーションの概念を紹介している。2つめの

> セクションは仮想化コンポーネント向けのセキュリティアドバイスを提供し、

> 3つめでは仮想化のプラニングと導入のセキュリティ確保について学ぶ。"

> "アドバイスはかなり一般的な内容で、ハイパーバイザーに関するものは以下の

> ようになっている。

> (中略)

> とはいえ、このガイドは仮想インフラの安全を確保するのに良い出発点となり、

> 新しいVMware vSphere 4.0 Security Hardening Guide」（vSphere 4.0セキュリ

> ティハードニングガイド）など、仮想化ベンダー各社がリリースしている具体的

> なハードニングガイドと合わせて読みたい。"

> "白書：vSphere 4.0 Security Hardening Guide（20100423-1）"

> http://virtualization.info/jp/news/2010/04/vsphere-40-security-hardening.html

> "vSphere 4.0 Security Hardening Guide"

> Created on: Apr 13, 2010 3:52 PM by Charu - Last Modified: Jul 7, 2010

> http://communities.vmware.com/docs/DOC-12306

"vSphere 4.1 Hardening Guide"

作成日: 2011/04/05 21:25 by Charu - 最終更新日: 2011/04/06 6:07 by Charu

http://communities.vmware.com/docs/DOC-15413

vSphere 5.0版は未だですね。

参考

"白書：VMware vSphere 5.0およびvCloud Director 1.5（20110802-1）"

August 3rd, 2011, Virtualization.info

http://virtualization.info/jp/news/2011/08/papers-vmware-vsphere-5-0-and-vcloud-director-1-5-html.html

Runnymede, England, 2009/10 [4]

2011-08-06T21:54:00.001+09:00

Runnymede, England, 2009/10 [1]

Runnymede, England, 2009/10 [2]
Runnymede, England, 2009/10 [3]

--------------------------------------------------

Runnymede

http://en.wikipedia.org/wiki/Runnymede

This may interest you

[san-tech][03411] 2012/01/25：京速コンピュータ「京」と創薬・医療の産学連携セミナー

[Slides/Papers]: Panasas, Object-Based Storage Device, Cray DVS, Berkeley RAID, Sprite OS Project (Log-Structured File Systems), InCast and etc.

[san-tech][03429] Re: PLFS, a Parallel Log-structured File System, Version2..1, 2011-10-13

[san-tech][03428] Slide: 5th Extremely Large Databases Conference (2011/10/18, 19)

[san-tech][03427] Re: International Magnetic Tape Storage Roadmap, Part I: Applications & Systems (2011/11/30)

[san-tech][03425] International Magnetic Tape Storage Roadmap, Part I: Applications & Systems (2011/11/30)

[san-tech][03424] Slide: 22nd Machine Evaluation Workshop (2011/11/29, 30)

[san-tech][03423] "NILFS 2.0.22 was released", Jun 19, 2011 JST.

[san-tech][03422] PLFS, a Parallel Log-structured File System, Version 2.1, 2011-10-13

[san-tech][03421] Research Note: Performance Modeling, Benchmarking and Simulation of High Performance Computer Systems (PMBS11)

[san-tech][03420] Paper: 1st Workshop on High-Performance Computing meets Databases at SC '11(2011/11/18)

[san-tech][03419] "2011 Thermal Guidelines for Liquid Cooled Data Processing Environments", ASHRAE Technical Committee (TC) 9.9

[san-tech][03418] Slide, Paper: 6th Parallel Data Storage Workshop at SC '11 (2011/11/13)

[san-tech][03417] Slide: OpenFabrics Alliance (OFA) at Supercomputing 11 (SC '11)

[san-tech][03416] IBM GPFS Native RAID & USENIX LISA '11

[san-tech][03415] pNFS in Technology Preview, Red Hat Enterprise Linux 6.2

[san-tech][03414] TMS introduces the High Availability, High Performance RamSan-720 (12-TB, SLC, 1U), December 6, 2011

[san-tech][03413] Re: Slide: HPC Advisory Council Stanford Workshop 2011 (2011/12/06, 07)

[san-tech][03410] Slide: October 2011 Open Networking Summit (2011/10/17 - 19)

[san-tech][03409] "日立が実証したデータセンタ冷却方式の選定基準が ITU（国際電気通信連合）の国際規格として採用"、2012/12/02

[san-tech][03407] Slide: Server Design Summit 2011 (2011/11/29, 30)

[san-tech][03405] Re: 東大京大でスパコンメーカ交代

[san-tech][03398] "OCZ Technology Acquires UK Design Team from PLX Technology", 2011-10-05

[san-tech][03396] "Red Hat to Acquire Gluster", October 4, 2011

[san-tech][03395] Fusion-io: ioDrive2 and ioDrive2 Duo, Oct. 3, 2011

Three "Bolshoi" Supercomputer Simulations of the Evolution of the Universe, Sep. 29, 2011

[san-tech][03394] Violin Memory: Violin 6000 Series & Violin 3220, Sep. 27, 2011

[san-tech][03393] QLogic: Highest Density, Most Flexible FC/FCoE Top of Rack (TOR) Switch, Sep 28, 2011

[san-tech][03391] DDN: SFA10Kシリーズ強化 (SFA10K-X & SFA10K-E), 09.20.11

[san-tech][03390] NetApp: 55PB (1TB/sec) Storage for LLNL Sequoia Supercomputer (20PF)'s Lustre, Sep. 28, 2011

[san-tech][03389] "Lustre 2.1 Released", September 30th 2011, Whamcloud Press Release

[san-tech][03388] Slide: European Lustre Workshop 2011 (2011/09/26, 27)

[san-tech][03386] "SDSC Announces Scalable, High-Performance Data Storage Cloud", 09.22.11

[san-tech][03385] Slide: HEC FSIO 2011 Workshop (2011/0807 - 10)

"DOD Announces First Strategy for Operating in Cyberspace", July 14, 2011

[san-tech][03382] OpenStackの大容量ストレージサービス、Swiftの仕組みと使い方、2011/08/05, 09/20, TechTarget

[san-tech][03381] Intel Hybrid Memory Cube (HMC) に関する Blog, Sep. 15, 2011, Research@Intel

[san-tech][03379] Cache IQ：新興 NAS加速化キャッシュプラットフォーム, Sep. 19, 2011

[san-tech][03378] ScaleMP Announces Strategic Collaboration with AMD, Sep. 20, 2011

[san-tech][03377] デル、最高45℃での使用も保証で外気冷却を支援、2011/08/24、＠IT

[san-tech][03375] Pure Storage: the First All-Flash Enterprise Array, August 23, 2011

[san-tech][03374] "Kaminario and Fusion-io Collaborate to Deliver Enterprise-class All Solid-State SAN Storage", Sept. 13, 2011

[san-tech][03373] Quantum to Expand StorNext Appliance Family, Sep 08, 2011

[san-tech][03372] IBM 100 petaOPS Supercomputer出願特許公開、September 8, 2011

[san-tech][03371] Broadcom to Acquire NetLogic Microsystems, September 12, 2011

[san-tech][03369] 小型のシリコンフォトニクス光源を開発、2011/09/16、富士通

[san-tech][03368] 2011/10/28 (金)：インテルソフトウェアカンファレンス東京

[san-tech][03367] "Koomey's Law: Computing efficiency keeps pace with Moore's Law", SEPTEMBER 13, 2011, InfoWorld

[san-tech][03366] HDS Announces Acquisition of BlueArc, September 7, 2011

[san-tech][03364] Texas Memory: 10 TB of eMLC Flash in 1U, August 23, 2011

[san-tech][03363] Fusion-io ioCache = ioDrive + ioTurbine, Aug. 29, 2011

[san-tech][03362] Citrix: Cloud.comをオープンソース化、8/29/2011

[san-tech][03361] Re: Slide: Flash Memory Summit 2011 Proceedings (2011/08/09 - 11)

[san-tech][03360] Intel Cilk Plus is now available in open-source and for GCC 4.7!, August 9, 2011

[san-tech][03359] The SSD Review (SSD情報サイト)

[san-tech][03358] Panasas Webinar: "pNFS-Are We There Yet?" (Aug. 24, 2011)

the human genome, Nature, Vol. 409, No. 6822, 15 February 2001

"Artificial Life", The Proceedings, September, 1987, etc.

[san-tech][03356] 白色LEDで 100Mbpsデータ転送 (100Mbps by white-light LEDs)

[san-tech][03354] Gluster Announces Apache Hadoop Storage Compatibility, August 23rd, 2011

[san-tech][03352] Slide: Titan Summit (Cray XK6, 20PF Peek), Oak Ridge Leadership Computing Facility (OLCF), (2011/08/15 - 17)

[san-tech][03351] Slide: US DOE Advanced CompuTational Software (ACTS) Collection (2011/08/16 - 19)

スーパーマイクロプロセッサ、bit 臨時増刊、1991年 3月

[san-tech][03350] Slide: US DOE 2011 Workshop on Architectures II: Exascale and Beyond: Configuring, Reasoning, Scaling (2011/08/08 - 10)

[san-tech][03349] Slide: US DOE 2011 Workshop on Architectures I: Exascale and Beyond: Gaps in Research, Gaps in our Thinking (2011/08/02, 03)

[san-tech][03346] Slide: 2011 Gordon Summer Institute (2011/08/08 - 12)

[san-tech][03345] Xsigo Server Fabric, August 15, 2011

Black & Red [02]

[san-tech][03344] SmartOS from Joyent (ZFS, DTrace, KVM (OS-based virtualization (Zones)), Crossbow)

[san-tech][03299] 2011/08/25(木)：SS研 HPCフォーラム2011 「ペタスケール HPC技術が支える地球科学」

[san-tech][03342] 次世代スーパーコンピュータ「京」ビデオ、ディスカバリーチャンネル＆BS-TBS

[san-tech][03341] Slide: Flash Memory Summit 2011 (2011/08/09 - 11)

[san-tech][02826] Virident tachIOn (PCIe SSD)：住友情報システムが代理店契約

2011/09/10 (土) 「三好甫先生記念計算科学シンポジウム」

Take a Bath!

Laleham Park, England, 2009/11

[san-tech][03339] SIGGRAPH 2011でも Fusion-io, Aug. 8, 2011

[san-tech][03337] IBM: NCSA Blue Watersキャンセルだ！, Aug. 8, 2011, Effective August 6, 2011

[san-tech][03336] New EMC Disk Library for Mainframe, The industry’s fastest mainframe VTL for z/OS, August 8, 2011

[san-tech][03299] 2011/08/25(木)：SS研 HPCフォーラム2011　「ペタスケール HPC技術が支える地球科学」