Background and Objectives

The PARKBENCH (PARallel Kernels and BENCHmarks) committee, originally called the Parallel Benchmark Working Group, PBWG, was founded at Supercomputing'92 in Minneapolis, when a group of about 50 people interested in computer benchmarking met under the joint initiative of Tony Hey (University of Southampton, UK) and Jack Dongarra (University of Tennessee/Oak Ridge National Laboratory). Most of the key players were present, from the Universities, Laboratories and industries, representing both computer manufacturers and computer users from both sides of the Atlantic. Roger Hockney (University of Southampton) chaired the meeting, and the objectives of the group were:

1. To establish a comprehensive set of parallel benchmarks that is generally accepted by both users and vendors of parallel system.

2. To provide a focus for parallel benchmark activities and avoid unnecessary duplication of effort and proliferation of benchmarks.

3. To set standards for benchmarking methodology and result-reporting together with a control database/repository for both the benchmarks and the results.

4. To make the benchmarks and results freely available in the public domain.

The first year's work was to produce a report and an initial set of benchmarks for release at Supercomputing'93 in Portland, Oregon, November 1993. The committee has met at the University of Tennessee at Knoxville on March 1--2, 1993, May 24, 1993 and August 23, 1993 to discuss the evolving draft of this report. The document reproduced here is the final result of these meetings, and is the first official publication of the PARKBENCH committee. It was distributed at a public Birds of a Feather meeting at Supercomputing'93, Portland, on 17th November 1993, as a University of Tennessee Technical Report CS-93-213 [1]. The bulk of this publication in Scientific Programming differs only in non-substantive editorial ways from the technical report. The first release of the PARKBENCH parallel benchmarks is available publicly over Internet.

The initial focus of the parallel benchmarks is on the new generation of scalable distributed-memory message-passing architectures for which there is a notable lack of existing benchmarks. For this reason the initial benchmark release concentrates on Fortran77 message-passing codes using the widely available PVM [2] message passing interface for portability. Future versions will undoubtedly adopt the proposed MPI [3] interface, when this is fully defined and becomes generally accepted. The committee's aim, however, is to cover all parallel architectures, and this is expected to be achieved by producing versions of the benchmark codes using Fortran90 and High Performance Fortran (HPF). Many shared-memory architectures provide efficient native implementations of PVM message-passing and are planning HPF compilers. They will be covered by these routes.