mpurtell@iastate.edu (Purtell Michael J) (01/29/91)
Archive-name: benchmark/supercomputer/slalom/1991-01-25 Archive-directory: tantalus.al.iastate.edu:/pub/Slalom/ [129.186.200.15] Original-posting-by: mpurtell@iastate.edu (Purtell Michael J) Original-subject: Re: SLALOM benchmarks Reposted-by: emv@ox.com (Edward Vielmetti) In article <1991Jan24.232704.28657@mstr.hgc.edu> keeg6544@sa8.hgc.edu.UUCP (michael keegan) writes: > >I am looking for any sources of information concerning the SLALOM benchmark. > >What recent papers/articles/reports have been published? Article in the December issue of Supercomputing Review. 'The Design of a Scalable, Fixed-Time Computer Benchmark' IS-5049/UC-32 by John Gustafson, Diane Rover, Stephen Elbert, and Michael Carter -- Ames Lab, Iowa State University >Have the highly parallel machines(CM,NCUBE,etc.) been benchmarked yet and if >so what were the results? Yes, results can be found in both of the above. For nCUBE 2, 20MHz using Fortran & tuned subroutine: processors Patches Operations MFlops 16 840 687.0M 11.5 64 1438 2.83G 47.2 256 2071 8.90G 153.0 For MasPar MP-1, 12.5MHz using C with plural variables (mpl): 1024 927 903.0M 15.9 2048 1055 1.30G 22.1 4096 1343 2.57G 43.7 8192 1663 4.73G 80.9 16384 2044 8.57G 144.0 >I'd also like to get some information/comments/trivia about a comparison >between the evaluation techniques used in the SLALOM and PERFECT benchmarks. Tantalus.al.iastate.edu has the SLALOM sources in various languages and other things including the latest reported results in pub/Slalom/Reports/CurrentReport >Mike Keegan keeg6544@mstr.hgc.edu -- Michael Purtell mpurtell@iastate.edu -- ------------------------------------------------------------------------------ Michael Purtell ---- mpurtell@iastate.edu Iowa State University ---- Birthplace of the first electronic digital computer ------------------------------------------------------------------------------