Skip to content
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.

Sign up for GitHub

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

Jump to bottom

Add Hyper Threading benchmarks workflow #1711

Merged
merged 12 commits into from
Apr 16, 2024
Merged

Add Hyper Threading benchmarks workflow #1711

merged 12 commits into from
Apr 16, 2024

Conversation

pefontana
Copy link
Collaborator

Add Hyper Threading benchmarks workflow

Description

Add a workflow to benchmark the Cairo-vm in a Hyper-Threading context

Checklist

  • Linked to Github Issue
  • Unit tests added
  • Integration tests added.
  • This change requires new documentation.
    • Documentation has been added/updated.
    • CHANGELOG has been updated.

@pefontana pefontana added the pipelines This PR/issue is exclusively about improving our CI label Apr 12, 2024
@github-actions GitHub Actions
Copy link

github-actions bot commented Apr 12, 2024 •
edited
Loading 

**Hyper Thereading Benchmark results**




hyperfine -r 2 -n "hyper_threading_main threads: 1" 'RAYON_NUM_THREADS=1 ./hyper_threading_main' -n "hyper_threading_pr threads: 1" 'RAYON_NUM_THREADS=1 ./hyper_threading_pr'
Benchmark 1: hyper_threading_main threads: 1
  Time (mean ± σ):     29.270 s ±  0.046 s    [User: 28.600 s, System: 0.666 s]
  Range (min … max):   29.237 s … 29.302 s    2 runs
 
Benchmark 2: hyper_threading_pr threads: 1
  Time (mean ± σ):     29.345 s ±  0.039 s    [User: 28.648 s, System: 0.694 s]
  Range (min … max):   29.318 s … 29.373 s    2 runs
 
Summary
  'hyper_threading_main threads: 1' ran
    1.00 ± 0.00 times faster than 'hyper_threading_pr threads: 1'




hyperfine -r 2 -n "hyper_threading_main threads: 2" 'RAYON_NUM_THREADS=2 ./hyper_threading_main' -n "hyper_threading_pr threads: 2" 'RAYON_NUM_THREADS=2 ./hyper_threading_pr'
Benchmark 1: hyper_threading_main threads: 2
  Time (mean ± σ):     15.714 s ±  0.021 s    [User: 28.777 s, System: 0.744 s]
  Range (min … max):   15.699 s … 15.729 s    2 runs
 
Benchmark 2: hyper_threading_pr threads: 2
  Time (mean ± σ):     15.741 s ±  0.031 s    [User: 28.914 s, System: 0.674 s]
  Range (min … max):   15.719 s … 15.763 s    2 runs
 
Summary
  'hyper_threading_main threads: 2' ran
    1.00 ± 0.00 times faster than 'hyper_threading_pr threads: 2'




hyperfine -r 2 -n "hyper_threading_main threads: 4" 'RAYON_NUM_THREADS=4 ./hyper_threading_main' -n "hyper_threading_pr threads: 4" 'RAYON_NUM_THREADS=4 ./hyper_threading_pr'
Benchmark 1: hyper_threading_main threads: 4
  Time (mean ± σ):     11.217 s ±  0.203 s    [User: 40.129 s, System: 0.907 s]
  Range (min … max):   11.073 s … 11.360 s    2 runs
 
Benchmark 2: hyper_threading_pr threads: 4
  Time (mean ± σ):     11.407 s ±  0.376 s    [User: 40.526 s, System: 0.934 s]
  Range (min … max):   11.141 s … 11.673 s    2 runs
 
Summary
  'hyper_threading_main threads: 4' ran
    1.02 ± 0.04 times faster than 'hyper_threading_pr threads: 4'




hyperfine -r 2 -n "hyper_threading_main threads: 6" 'RAYON_NUM_THREADS=6 ./hyper_threading_main' -n "hyper_threading_pr threads: 6" 'RAYON_NUM_THREADS=6 ./hyper_threading_pr'
Benchmark 1: hyper_threading_main threads: 6
  Time (mean ± σ):     11.056 s ±  0.040 s    [User: 40.792 s, System: 0.923 s]
  Range (min … max):   11.027 s … 11.085 s    2 runs
 
Benchmark 2: hyper_threading_pr threads: 6
  Time (mean ± σ):     11.341 s ±  0.083 s    [User: 40.571 s, System: 0.945 s]
  Range (min … max):   11.282 s … 11.400 s    2 runs
 
Summary
  'hyper_threading_main threads: 6' ran
    1.03 ± 0.01 times faster than 'hyper_threading_pr threads: 6'




hyperfine -r 2 -n "hyper_threading_main threads: 8" 'RAYON_NUM_THREADS=8 ./hyper_threading_main' -n "hyper_threading_pr threads: 8" 'RAYON_NUM_THREADS=8 ./hyper_threading_pr'
Benchmark 1: hyper_threading_main threads: 8
  Time (mean ± σ):     10.978 s ±  0.346 s    [User: 40.818 s, System: 0.986 s]
  Range (min … max):   10.733 s … 11.222 s    2 runs
 
Benchmark 2: hyper_threading_pr threads: 8
  Time (mean ± σ):     11.066 s ±  0.027 s    [User: 41.438 s, System: 0.901 s]
  Range (min … max):   11.047 s … 11.085 s    2 runs
 
Summary
  'hyper_threading_main threads: 8' ran
    1.01 ± 0.03 times faster than 'hyper_threading_pr threads: 8'




hyperfine -r 2 -n "hyper_threading_main threads: 16" 'RAYON_NUM_THREADS=16 ./hyper_threading_main' -n "hyper_threading_pr threads: 16" 'RAYON_NUM_THREADS=16 ./hyper_threading_pr'
Benchmark 1: hyper_threading_main threads: 16
  Time (mean ± σ):     11.301 s ±  0.225 s    [User: 41.146 s, System: 0.998 s]
  Range (min … max):   11.142 s … 11.460 s    2 runs
 
Benchmark 2: hyper_threading_pr threads: 16
  Time (mean ± σ):     11.163 s ±  0.054 s    [User: 42.116 s, System: 0.985 s]
  Range (min … max):   11.125 s … 11.201 s    2 runs
 
Summary
  'hyper_threading_pr threads: 16' ran
    1.01 ± 0.02 times faster than 'hyper_threading_main threads: 16'

@codecov Codecov
Copy link

codecov bot commented Apr 12, 2024 •
edited
Loading

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 94.90%. Comparing base (e629428) to head (a005262).

Additional details and impacted files 
@@           Coverage Diff           @@
##             main    #1711   +/-   ##
=======================================
  Coverage   94.90%   94.90%           
=======================================
  Files          98       98           
  Lines       38092    38092           
=======================================
  Hits        36151    36151           
  Misses       1941     1941

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

@github-actions GitHub Actions
Copy link

github-actions bot commented Apr 12, 2024 •
edited
Loading

Benchmark Results for unmodified programs 🚀

Command Mean [s] Min [s] Max [s] Relative
base big_factorial 2.299 ± 0.022 2.282 2.357 1.00 ± 0.01
head big_factorial 2.297 ± 0.020 2.277 2.345 1.00
Command Mean [s] Min [s] Max [s] Relative
base big_fibonacci 2.287 ± 0.019 2.264 2.326 1.00
head big_fibonacci 2.314 ± 0.055 2.276 2.454 1.01 ± 0.03
Command Mean [s] Min [s] Max [s] Relative
base blake2s_integration_benchmark 8.584 ± 0.171 8.364 8.732 1.00
head blake2s_integration_benchmark 8.695 ± 0.104 8.406 8.775 1.01 ± 0.02
Command Mean [s] Min [s] Max [s] Relative
base compare_arrays_200000 2.364 ± 0.069 2.327 2.557 1.00 ± 0.03
head compare_arrays_200000 2.362 ± 0.024 2.333 2.413 1.00
Command Mean [s] Min [s] Max [s] Relative
base dict_integration_benchmark 1.472 ± 0.011 1.456 1.488 1.00 ± 0.01
head dict_integration_benchmark 1.468 ± 0.012 1.453 1.487 1.00
Command Mean [s] Min [s] Max [s] Relative
base field_arithmetic_get_square_benchmark 1.397 ± 0.012 1.380 1.418 1.00
head field_arithmetic_get_square_benchmark 1.413 ± 0.019 1.391 1.441 1.01 ± 0.02
Command Mean [s] Min [s] Max [s] Relative
base integration_builtins 8.636 ± 0.273 8.347 9.285 1.02 ± 0.04
head integration_builtins 8.459 ± 0.158 8.340 8.692 1.00
Command Mean [s] Min [s] Max [s] Relative
base keccak_integration_benchmark 8.785 ± 0.169 8.635 9.001 1.00 ± 0.03
head keccak_integration_benchmark 8.761 ± 0.187 8.602 8.999 1.00
Command Mean [s] Min [s] Max [s] Relative
base linear_search 2.383 ± 0.021 2.356 2.421 1.00 ± 0.02
head linear_search 2.383 ± 0.032 2.353 2.457 1.00
Command Mean [s] Min [s] Max [s] Relative
base math_cmp_and_pow_integration_benchmark 1.834 ± 0.021 1.812 1.871 1.00 ± 0.01
head math_cmp_and_pow_integration_benchmark 1.826 ± 0.015 1.802 1.847 1.00
Command Mean [s] Min [s] Max [s] Relative
base math_integration_benchmark 1.617 ± 0.015 1.599 1.643 1.00
head math_integration_benchmark 1.636 ± 0.018 1.611 1.670 1.01 ± 0.01
Command Mean [s] Min [s] Max [s] Relative
base memory_integration_benchmark 1.283 ± 0.015 1.269 1.317 1.00
head memory_integration_benchmark 1.287 ± 0.009 1.273 1.299 1.00 ± 0.01
Command Mean [s] Min [s] Max [s] Relative
base operations_with_data_structures_benchmarks 1.922 ± 0.006 1.913 1.929 1.00
head operations_with_data_structures_benchmarks 1.933 ± 0.024 1.920 2.000 1.01 ± 0.01
Command Mean [ms] Min [ms] Max [ms] Relative
base pedersen 547.1 ± 1.7 545.0 550.3 1.00
head pedersen 548.7 ± 4.2 544.0 559.1 1.00 ± 0.01
Command Mean [ms] Min [ms] Max [ms] Relative
base poseidon_integration_benchmark 977.6 ± 8.0 969.2 996.9 1.00
head poseidon_integration_benchmark 982.8 ± 8.5 968.9 996.3 1.01 ± 0.01
Command Mean [s] Min [s] Max [s] Relative
base secp_integration_benchmark 1.928 ± 0.021 1.914 1.978 1.00 ± 0.01
head secp_integration_benchmark 1.926 ± 0.015 1.908 1.962 1.00
Command Mean [ms] Min [ms] Max [ms] Relative
base set_integration_benchmark 746.0 ± 2.4 743.3 751.4 1.00
head set_integration_benchmark 748.1 ± 5.0 742.4 758.1 1.00 ± 0.01
Command Mean [s] Min [s] Max [s] Relative
base uint256_integration_benchmark 4.685 ± 0.051 4.643 4.763 1.00
head uint256_integration_benchmark 4.711 ± 0.056 4.639 4.763 1.01 ± 0.02

Oppen
Oppen reviewed Apr 15, 2024
View reviewed changes
.github/workflows/hyper_threading_benchmarks.yml Outdated
Comment on lines 62 to 84
- name: Upload Main Binary
uses: actions/upload-artifact@v4
with:
name: hyper_threading_main_binary
path: ${{ github.workspace }}/hyper_threading_main

- name: Download hyper_threading_pr_binary
uses: actions/download-artifact@v4
with:
name: hyper_threading_pr_binary
path: ${{ github.workspace }}/

- name: Download hyper_threading_workflow_script
uses: actions/download-artifact@v4
with:
name: hyper_threading_workflow_script
path: ${{ github.workspace }}/

- name: Download hyper_threading_main_binary
uses: actions/download-artifact@v4
with:
name: hyper_threading_main_binary
path: ${{ github.workspace }}/
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

This is unnecessary. Since you're running everything in a single job you still have the files where you copied them. This would only be necessary if we did parallel jobs here.

Copy link
Collaborator Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Thank @Oppen !!
Since we are doing a git checkout main (To create the other binary and compare it vs the branch one) I still need to download and upload the PR binary and the workflow.sh. But I was able to remove the download and upload of the main binary
0726f23

Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Right. We may be able to remove that too in a later PR, when this is in main.

Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Hold on. We can make the copy to the workspace before checking out main. That should solve it.

Oppen
Oppen reviewed Apr 15, 2024
View reviewed changes
.github/workflows/hyper_threading_benchmarks.yml
Comment on lines +39 to +51
- name: Upload PR Binary
uses: actions/upload-artifact@v4
with:
name: hyper_threading_pr_binary
path: ${{ github.workspace }}/hyper_threading_pr

- name: Upload Workflow Script
uses: actions/upload-artifact@v4
with:
name: hyper_threading_workflow_script
path: ${{ github.workspace }}/hyper-threading-workflow.sh


Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Same for this.

@pefontana pefontana requested a review from Oppen April 15, 2024 21:12
@fmoletta fmoletta added this pull request to the merge queue Apr 16, 2024
@fmoletta fmoletta removed this pull request from the merge queue due to a manual request Apr 16, 2024
@pefontana pefontana enabled auto-merge April 16, 2024 21:48
@pefontana pefontana added this pull request to the merge queue Apr 16, 2024
Merged via the queue into main with commit 584eaa2 Apr 16, 2024
71 checks passed
@pefontana pefontana deleted the add-hyper-threading-workflow branch April 16, 2024 22:35
@pefontana pefontana linked an issue Apr 17, 2024 that may be closed by this pull request
Add hyper threading benchmarks to the CI #1675
Closed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers

@juanbono juanbono juanbono approved these changes

@fmoletta fmoletta fmoletta approved these changes

@igaray igaray Awaiting requested review from igaray igaray is a code owner

@Oppen Oppen Awaiting requested review from Oppen Oppen is a code owner

Assignees
No one assigned
Labels
pipelines This PR/issue is exclusively about improving our CI
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.

Add hyper threading benchmarks to the CI
4 participants
@pefontana @Oppen @juanbono @fmoletta