PARKBENCH 99 - Related Links
The NAS Parallel Benchmarks (NPB) are a set of 8 programs designed to help evaluate the performance of parallel supercomputers. The benchmarks, which are derived from computational fluid dynamics (CFD) applications, consist of five kernels and three pseudo-applications. The NPB come in two flavors:
These are the original "pencil and paper" benchmarks. Vendors and others implement the detailed specifications in the NPB 1 report, using algorithms and programming models appropriate to their different machines. The results are verified by NAS and published in a periodic NAS report. NPB 1 implementations are generally proprietary.
These are MPI-based source-code implementations written and distributed by NAS. They are intended to be run with little or no tuning, and approximate the performance a typical user can expect to obtain for a portable parallel program. They supplement, rather than replace, NPB 1. NAS solicits performance results from all sources.
These are single processor (serial) source-code implementations derived from the NPB 2 by removing all parallelism. They are intended to be starting points for shared memory and distributed memory versions, as tests for parallelization tools, and as benchmarks for workstations and PC's (the new Class W size). As benchmarks, they are intended to be run with little or no tuning. NAS solicits performance results from all sources.
The benchmark subdirectory of this site contains the original Linpack benchmark used to rank the top 500 Supercomputer Sites.
RAPS is a European Software Initiative developing a benchmark suite for parallel computers based on realistic applications. Many European HPC users and leading HPC vendors cooperate in this activity.
Section 1a: Low Level Operations
Generic Code Segments
Section 2a: Kernels
Tower of Hanoi
Sieve of Eratosthenes for finding prime numbers
2-D Fast Fourier Transform
NAS EP benchmark generating Gaussian random numbers
NAS IS benchmark for integer sorting
The DHPC are currently working on implementing Java versions of other
Parallel and Distributed Computing
Department of Electronics & Computer Science
University of Southampton