Skip to content

feat: measure the RAM footprint of benchmarks by using /usr/bin/time util #31

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
wants to merge 21 commits into
base: CSP-Q3-2025
Choose a base branch
from

Conversation

guorong009
Copy link
Collaborator

@guorong009 guorong009 commented Aug 18, 2025

Description

  • Add the shell script for using the /usr/bin/time util in measuring the RAM footprint of benchmarks
  • Refactor the relating benchmarks

Changes

  • add measure_mem_avg.sh script
    • it can set number of runs(e.g NUM_RUNS=4 ../measure_mem_avg.sh ...; default value: 10)
    • it can also set the output json file(e.g ../measure_mem_avg.sh --json sha2_binius_mem_report.json -- [CMD])
  • refactor the binius, plonky2, plonky3-powdr, provekit benchmarks
    • move the SubMetrics measurement to criterion bench code
    • refactor the provekit benchmark

NOTE

  • omit the plonky3-sp1 since it could be replaced by ere

@guorong009 guorong009 requested review from alxkzmn and brech1 August 18, 2025 10:45
@guorong009 guorong009 marked this pull request as ready for review August 18, 2025 10:45
@alxkzmn
Copy link
Member

alxkzmn commented Aug 19, 2025

@guorong009 I see that you retained the RAM measurement in the old "measure" binaries, but I think it's ok to delete them and move the proof/preprocessing size measurement into the criterion benchmarks before the timings measurement.

@guorong009 guorong009 marked this pull request as draft August 19, 2025 09:06
@guorong009 guorong009 marked this pull request as ready for review August 20, 2025 02:42
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

3 participants