HPFBench: A High Performance Fortran benchmark suite

Citation
Yc. Hu et al., HPFBench: A High Performance Fortran benchmark suite, ACM T MATH, 26(1), 2000, pp. 99-149
Citations number
52
Categorie Soggetti
Computer Science & Engineering
Journal title
ACM TRANSACTIONS ON MATHEMATICAL SOFTWARE
ISSN journal
00983500 → ACNP
Volume
26
Issue
1
Year of publication
2000
Pages
99 - 149
Database
ISI
SICI code
0098-3500(200003)26:1<99:HAHPFB>2.0.ZU;2-E
Abstract
The High Performance Fortran (HPF) benchmark suite HPFBench is designed for evaluating the HPF language and compilers on scalable architectures. The f unctionality of the benchmarks covers scientific software library functions and application kernels that reflect the computational structure and commu nication patterns in fluid dynamic simulations, fundamental physics, and mo lecular studies in chemistry and biology. The benchmarks are characterized in terms of FLOP count, memory usage, communication pattern, local memory a ccesses, array allocation mechanism, as well as operation and communication counts per iteration. The benchmarks output performance evaluation metrics in the form of elapsed times, FLOP rates, and communication time breakdown s. We also provide a benchmark guide to aid the choice of subsets of the be nchmarks for evaluating particular aspects of an HPF compiler. Furthermore, we report an evaluation of an industry-leading HPF compiler from the Portl and Group Inc. using the HPFBench benchmarks on the distributed-memory IBM SP2.