Progress on H5Part: A Portable High Performance Parallel DataInterface for Electromagnetics Simulations

PDF Version Also Available for Download.

Description

Significant problems facing all experimental andcomputationalsciences arise from growing data size and complexity. Commonto allthese problems is the need to perform efficient data I/O ondiversecomputer architectures. In our scientific application, thelargestparallel particle simulations generate vast quantitiesofsix-dimensional data. Such a simulation run produces data foranaggregate data size up to several TB per run. Motived by the needtoaddress data I/O and access challenges, we have implemented H5Part,anopen source data I/O API that simplifies the use of the HierarchicalDataFormat v5 library (HDF5). HDF5 is an industry standard forhighperformance, cross-platform data storage and retrieval that runsonall contemporary architectures from large parallel supercomputerstolaptops. H5Part, which ... continued below

Creation Information

Adelmann, Andreas; Gsell, Achim; Oswald, Benedikt; Schietinger,Thomas; Bethel, Wes; Shalf, John et al. June 22, 2007.

Context

This article is part of the collection entitled: Office of Scientific & Technical Information Technical Reports and was provided by UNT Libraries Government Documents Department to Digital Library, a digital repository hosted by the UNT Libraries. More information about this article can be viewed below.

Who

People and organizations associated with either the creation of this article or its content.

Provided By

UNT Libraries Government Documents Department

Serving as both a federal and a state depository library, the UNT Libraries Government Documents Department maintains millions of items in a variety of formats. The department is a member of the FDLP Content Partnerships Program and an Affiliated Archive of the National Archives.

Contact Us

What

Descriptive information to help identify this article. Follow the links below to find similar items on the Digital Library.

Description

Significant problems facing all experimental andcomputationalsciences arise from growing data size and complexity. Commonto allthese problems is the need to perform efficient data I/O ondiversecomputer architectures. In our scientific application, thelargestparallel particle simulations generate vast quantitiesofsix-dimensional data. Such a simulation run produces data foranaggregate data size up to several TB per run. Motived by the needtoaddress data I/O and access challenges, we have implemented H5Part,anopen source data I/O API that simplifies the use of the HierarchicalDataFormat v5 library (HDF5). HDF5 is an industry standard forhighperformance, cross-platform data storage and retrieval that runsonall contemporary architectures from large parallel supercomputerstolaptops. H5Part, which is oriented to the needs of the particlephysicsand cosmology communities, provides support for parallelstorage andretrieval of particles, structured and in the future unstructuredmeshes.In this paper, we describe recent work focusing on I/O supportforparticles and structured meshes and provide data showing performance onmodernsupercomputer architectures like the IBM POWER 5.

Source

  • Particle Accelerator Conference 2007,Albuquerque, New Mexico, USA, June 25 - 29, 2007

Language

Item Type

Identifier

Unique identifying numbers for this article in the Digital Library or other systems.

  • Report No.: LBNL--63042
  • Grant Number: DE-AC02-05CH11231
  • Office of Scientific & Technical Information Report Number: 929325
  • Archival Resource Key: ark:/67531/metadc901858

Collections

This article is part of the following collection of related materials.

Office of Scientific & Technical Information Technical Reports

What responsibilities do I have when using this article?

When

Dates and time periods associated with this article.

Creation Date

  • June 22, 2007

Added to The UNT Digital Library

  • Sept. 27, 2016, 1:39 a.m.

Usage Statistics

When was this article last used?

Yesterday: 1
Past 30 days: 1
Total Uses: 6

Interact With This Article

Here are some suggestions for what to do next.

Start Reading

PDF Version Also Available for Download.

Citations, Rights, Re-Use

Adelmann, Andreas; Gsell, Achim; Oswald, Benedikt; Schietinger,Thomas; Bethel, Wes; Shalf, John et al. Progress on H5Part: A Portable High Performance Parallel DataInterface for Electromagnetics Simulations, article, June 22, 2007; United States. (digital.library.unt.edu/ark:/67531/metadc901858/: accessed August 18, 2017), University of North Texas Libraries, Digital Library, digital.library.unt.edu; crediting UNT Libraries Government Documents Department.