Published July 2007 | Version v1
Journal article Metadata-only

Performance and scaling of locally-structured grid methods for partial differential equations

Creators

  • 1. Computational Research Division, Lawrence Berkeley National Laboratory, 1 Cyclotron Road, Berkeley, CA 94720 (United States)

Description

In this paper, we discuss some of the issues in obtaining high performance for block-structured adaptive mesh refinement software for partial differential equations. We show examples in which AMR scales to thousands of processors. We also discuss a number of metrics for performance and scalability that can provide a basis for understanding the advantages and disadvantages of this approach

Additional details

Identifiers

DOI
10.1088/1742-6596/78/1/012013;

Publishing Information

Journal Title
Journal of Physics. Conference Series (Online)
Journal Volume
78
Journal Issue
1
Journal Page Range
p. 012013
ISSN
1742-6596

Conference

Title
Scientific Discovery through Advanced Computing
Acronym
SciDAC 2007
Dates
24-28 Jun 2007
Place
Boston, MA (United States)

INIS

Country of Publication
United Kingdom
Country of Input or Organization
International Atomic Energy Agency (IAEA)
INIS RN
39029211
Subject category
S71: CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSICS;
Resource subtype / Literary indicator
Conference
Descriptors DEI
CALCULATION METHODS; COMPUTER CALCULATIONS; COMPUTER CODES; PARTIAL DIFFERENTIAL EQUATIONS; PERFORMANCE
Descriptors DEC
DIFFERENTIAL EQUATIONS; EQUATIONS