Published July 2011 | Version v1
Journal article Metadata-only

High performance iterative solver for linear system using multi GPU

Creators

  • 1. Tokyo University of Technology, Hachioji, Tokyo (Japan)
  • 2. Seikei University, Musashino, Tokyo (Japan)
  • 3. Ritsumeikan University, Kusatsu, Shiga (Japan)
  • 4. Hokkaido University, Sapporo, Hokkaido (Japan)
  • 5. National Institute for Fusion Science, Toki, Gifu (Japan)

Description

The variable preconditioned (VP) Krylov subspace method on multi Graphics Processing Unit (GPU) is numerically investigated. Besides, the linear system obtained by finite element method with an edge element is adopted for the problem. The results of computations show that VP conjugate gradient method on multi GPU demonstrated significant achievement than that of CPU. Especially, VP conjugate gradient method on multi GPU is 4.35 times faster than that of CPU. However, transmission rate between the PC using Gigabit Ethernet is the bottleneck of the performance. (author)

Availability note (English)

Available from http://dx.doi.org/10.1585/pfr.6.2401069

Additional details

Identifiers

DOI
10.1585/pfr.6.2401069;

Publishing Information

Journal Title
Plasma and Fusion Research
Journal Volume
6
Journal Issue
special issue 1
Journal Page Range
p. 2401069.1-2401069.5
ISSN
1880-6821

Conference

Title
20. international Toki conference on the next twenty years in plasma and fusion science
Acronym
ITC20
Dates
7-10 Dec 2010
Place
Toki, Gifu (Japan)

INIS

Country of Publication
Japan
Country of Input or Organization
Japan
INIS RN
46103753
Subject category
S70: PLASMA PHYSICS AND FUSION TECHNOLOGY;
Resource subtype / Literary indicator
Conference
Descriptors DEI
BOUNDARY-VALUE PROBLEMS; COMPUTER ARCHITECTURE; COMPUTER CALCULATIONS; COMPUTER GRAPHICS; COMPUTER-GRAPHICS DEVICES; COMPUTERIZED SIMULATION; FINITE ELEMENT METHOD; ITERATIVE METHODS; PERFORMANCE TESTING; PLASMA SIMULATION
Descriptors DEC
CALCULATION METHODS; COMPUTER OUTPUT DEVICES; MATHEMATICAL SOLUTIONS; NUMERICAL SOLUTION; SIMULATION; TESTING

Optional Information

Notes
9 refs., 5 figs., 1 tab.