Skip to content

Commit

Permalink
Specify CUDA GPU for benchmarks
Browse files Browse the repository at this point in the history
  • Loading branch information
rkierulf committed Jul 1, 2024
1 parent 8f23b27 commit dc4b62a
Show file tree
Hide file tree
Showing 2 changed files with 2 additions and 6 deletions.
2 changes: 1 addition & 1 deletion .buildkite/runbenchmarks.yml
Original file line number Diff line number Diff line change
Expand Up @@ -78,7 +78,7 @@ steps:
- "bench/results/*"
agents:
queue: "benchmark"
gpu: "*"
gpu: "rtx2070"
cuda: "*"
env:
BENCHMARK_GROUP: CUDA
Expand Down
6 changes: 1 addition & 5 deletions .github/workflows/Benchmark.yml
Original file line number Diff line number Diff line change
Expand Up @@ -51,11 +51,7 @@ jobs:
- name: Locate Benchmarks Artifact
id: locate
if: ${{ steps.download.outputs.download-state == 'success' }}
run: |
pwd
ls
echo $(find artifacts -type f -name combinedbenchmarks.json)
echo "path=$(find artifacts -type f -name combinedbenchmarks.json 2>/dev/null)" >> $GITHUB_OUTPUT
run: echo "path=$(find artifacts -type f -name combinedbenchmarks.json 2>/dev/null)" >> $GITHUB_OUTPUT

- name: Upload Benchmark Results
if: ${{ steps.locate.outputs.path != '' }}
Expand Down

1 comment on commit dc4b62a

@github-actions
Copy link
Contributor

Choose a reason for hiding this comment

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

KomaMRI Benchmarks

Benchmark suite Current: dc4b62a Previous: 8f23b27 Ratio
MRI Lab/Bloch/CPU/2 thread(s) 1553639100.5 ns 1610218507 ns 0.96
MRI Lab/Bloch/CPU/4 thread(s) 921649522 ns 897695394 ns 1.03
MRI Lab/Bloch/CPU/8 thread(s) 831399922.5 ns 842009155 ns 0.99
MRI Lab/Bloch/CPU/1 thread(s) 3386467342.5 ns 3298254646 ns 1.03
MRI Lab/Bloch/GPU/CUDA 153247901 ns 156847184 ns 0.98
MRI Lab/Bloch/GPU/oneAPI 27765566695 ns 27199211288 ns 1.02
MRI Lab/Bloch/GPU/Metal 3420303833.5 ns 3364538500.5 ns 1.02
MRI Lab/Bloch/GPU/AMDGPU 1731801925.5 ns 1745536590 ns 0.99
Slice Selection 3D/Bloch/CPU/2 thread(s) 3656186710 ns 3737687010.5 ns 0.98
Slice Selection 3D/Bloch/CPU/4 thread(s) 1738954666 ns 1753092848.5 ns 0.99
Slice Selection 3D/Bloch/CPU/8 thread(s) 1248648438 ns 1191746688 ns 1.05
Slice Selection 3D/Bloch/CPU/1 thread(s) 5965079883.5 ns 6080984636 ns 0.98
Slice Selection 3D/Bloch/GPU/CUDA 265424409 ns 266744286 ns 1.00
Slice Selection 3D/Bloch/GPU/oneAPI 2171391289 ns 2168572006 ns 1.00
Slice Selection 3D/Bloch/GPU/Metal 1398666854.5 ns 1398475208 ns 1.00
Slice Selection 3D/Bloch/GPU/AMDGPU 759035177 ns 716831522 ns 1.06

This comment was automatically generated by workflow using github-action-benchmark.

Please sign in to comment.