Tera-scalable Algorithms for Variable-Density Elliptic Hydrodynamics with Spectral Accuracy

PDF Version Also Available for Download.

Description

A hybrid spectral/compact solver for variable-density viscous incompressible flow is described. Parallelization strategies for the FFTs and band-diagonal matrices are discussed and compared. Transpose methods are found to be highly competitive with direct block parallel methods when the problem is scaled to tens of thousands of processors. Various mapping strategies for the IBM BlueGene/L torus configuration of processors are explored. By optimizing the communication, we have achieved virtually perfect scaling to 32768 nodes. Furthermore, communication rates come very close to the theoretical peak speed of the BlueGene/L network with sustained computation in the TeraFLOPS range.

Physical Description

PDF-file: 14 pages; size: 0.8 Mbytes

Creation Information

Cook, A W; Cabot, W H; Welcome, M L; Williams, P L; Miller, B J; de Supinski, B R et al. April 13, 2005.

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.

Publisher

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

A hybrid spectral/compact solver for variable-density viscous incompressible flow is described. Parallelization strategies for the FFTs and band-diagonal matrices are discussed and compared. Transpose methods are found to be highly competitive with direct block parallel methods when the problem is scaled to tens of thousands of processors. Various mapping strategies for the IBM BlueGene/L torus configuration of processors are explored. By optimizing the communication, we have achieved virtually perfect scaling to 32768 nodes. Furthermore, communication rates come very close to the theoretical peak speed of the BlueGene/L network with sustained computation in the TeraFLOPS range.

Physical Description

PDF-file: 14 pages; size: 0.8 Mbytes

Source

  • Presented at: Supercomputing, Seattle, WA, United States, Nov 12 - Nov 18, 2005

Language

Item Type

Identifier

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

  • Report No.: UCRL-CONF-211384
  • Grant Number: W-7405-ENG-48
  • Office of Scientific & Technical Information Report Number: 877898
  • Archival Resource Key: ark:/67531/metadc880975

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

  • April 13, 2005

Added to The UNT Digital Library

  • Sept. 21, 2016, 2:29 a.m.

Description Last Updated

  • Dec. 2, 2016, 3:17 p.m.

Usage Statistics

When was this article last used?

Congratulations! It looks like you are the first person to view this item online.

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

Cook, A W; Cabot, W H; Welcome, M L; Williams, P L; Miller, B J; de Supinski, B R et al. Tera-scalable Algorithms for Variable-Density Elliptic Hydrodynamics with Spectral Accuracy, article, April 13, 2005; Livermore, California. (digital.library.unt.edu/ark:/67531/metadc880975/: accessed September 22, 2017), University of North Texas Libraries, Digital Library, digital.library.unt.edu; crediting UNT Libraries Government Documents Department.