Reprints from my posting to SAN-Tech Mailing List and ...


Adaptable IO System (ADIOS) 1.3 Released, July 12, 2011

Adaptable IO System (ADIOS) 
"ADIOS Middleware Update Sets the Stage for Exascale Computing"
 July 26, 2012, Oak Ridge Leadership Computing Facility

"Researchers can multitask and access files simultaneously with version 1.4"
1.4.0 Released July 2012

*) new read API for staging methods with step-by-step processing,
   also with non-blocking and chunking APIs
*) visualization schema added to ADIOS XML format
*) skel: I/O skeleton generator and evaluation tools
*) unified error/debug logging
*) hidden attributes (version, create/update times)
*) Java and Numpy bindings
*) F90 modules adios_write_mod and adios_read_mod to check syntax
   at compile time
Revision History
1.3 Released July 12, 2011

*)New read method BP_STAGED for reading files with improved performance
*)Changes in MPI_AMR and MPI_LUSTRE to support default parameters.
  Parameters for MPI_LUSTRE and MPI_AMR in XML file are not mandatory
  any more.

"ORNL ADIOS Team Releases Version 1.3 of Adaptive Input/Output System"
 July 21, 2011, HPCwire

  "What good are supercomputers if you have to spend your whole time
   getting data in and out? Researchers at Oak Ridge National Laboratory
   (ORNL) are working to say goodbye to input/output (I/O) problems
   with their most recent upgrade of the Adaptive Input/Output System
  "The open-source middleware is designed to help researchers maximize
   their allocations on leadership-class computing resources from
   wherever they may be. In essence, it creates more time for research
   by minimizing the time needed to read and write data to files,
   even if researchers are sending those files from thousands of miles
  "The biggest challenge for the newest release - version 1.3 - was
   to improve reading efficiency, according to Qing Liu, a scientific
   researcher working in the Remote Data Analysis and Visualization
   branch at the National Institute for Computational Sciences and
   the leading developer of ADIOS."

"As Supercomputers Approach Exascale, Experts Wrestle with Big Data"
 July 21, 2011, Dawn Levy, OLCF science writer, for HPCwire,_experts_wrestle_with_big_data.html

  "The Oak Ridge Leadership Computing Facility (OLCF) plans on delivering
   a system with peak performance of 10?20 petaflops at Oak Ridge National
   Laboratory (ORNL) in 2012. This system will begin the OLCF's transitionfrom
   petaflop to exaflop computing this decade. Providing an environment
   that balances computational speed, memory bandwidth, and the input/output
   of data - so no aspect bottlenecks the performance of scientific
   applications - will require dramatic advances in parallel I/O capabilities.
   These high performance computing systems will generate unprecedented
   amounts of data - and unprecedented challenges in managing it.

   In this interview, Galen Shipman, who heads OLCF's Technology Integration
   group, discusses the challenges of managing big data. Shipman's group is
   charged with integrating advanced technologies in the areas of networks,
   file systems, and archival storage infrastructures into the OLCF's HPC
   systems, such as Jaguar, a 2.33-petaflop Cray XT funded by the Department
   of Energy Office of Science."
[san-tech][02840] "Tuning HDF5 for Lustre File Systems", IASDS10 (2010/09/24)
  Date: Sun, 28 Nov 2010 06:42:20 +0900

0 件のコメント: