[gambit-list] Comparing performance of various machines on Gambit running array test program.

Lucier, Bradley J lucier at purdue.edu
Sun Mar 27 15:10:30 EDT 2022


I compared the run time of the test program of SRFI 231, an array SRFI:

https://srfi.schemers.org/srfi-231/

by upping the number of samples in each random test to 10,000 by setting random-tests to 10000 in

https://github.com/scheme-requests-for-implementation/srfi-231/blob/master/test-arrays.scm 

I used the same version of Gambit and gcc on all machines:

lucier% gsi -v
v4.9.4-26-g015c8875 20220323161655 x86_64-apple-darwin20.6.0 "./configure 'CC=/usr/local/gcc-11.2.0/bin/gcc' '--enable-march=native' '--enable-dynamic-clib' '--enable-single-host' '--enable-multiple-versions' '--enable-shared’”

The results:

Dell Workstation running Ubuntu 20.04

model name	: Intel(R) Xeon(R) CPU E3-1271 v3 @ 3.60GHz
144.279u 0.443s 2:25.12 99.7%	0+0k 8+1024io 0pf+0w
32GB RAM

  Model Identifier:	MacBookPro18,4
  Chip:	Apple M1 Max
Up to 3.228 GHz
64GB RAM
69.523u 0.324s 1:10.25 99.4%	0+0k 0+0io 0pf+0w

  Model Identifier:	Macmini8,1
  Processor Name:	6-Core Intel Core i7
  Processor Speed:	3.2 GHz
  Intel Core i7-8700B @ 3.2 GHz
107.773u 0.255s 1:48.65 99.4%   0+0k 0+0io 0pf+0w

So running this, the Mac mini is significantly faster than the Linux box, but the Linux box feels faster for nearly all development.

I’m quite pleased with the MacBook Pro, though.

Brad


More information about the Gambit-list mailing list