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.

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

[WIP] dgl-cugraph benchmarks #3078

Closed

Conversation

VibhuJawa
Copy link
Member

@VibhuJawa VibhuJawa commented Dec 13, 2022

This PR depends upon: #3048

Example Command:

pytest -svx --benchmark-columns="mean,stddev,outliers,rounds" --benchmark-gpu-disable --benchmark-warmup=off --benchmark-min-rounds=2 --pdb cugraph-dgl/pytest-based/bench_cugraph_dgl_uniform_neighbor_sample.py -k "SG and fanout_10_25 and rmat_25_4"

----------------------------------------------------------------------------------------- benchmark: 15 tests -----------------------------------------------------------------------------------------
Name (time in ms, mem in bytes)                                                                                                          Mean              StdDev            Outliers  Rounds          
-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
bench_cugraph_dgl_uniform_neighbor_sample[gpu_config=SG-dataset=rmat_25_4-with_replacement=False-fanout=10_25-batch_size=100]         83.6024 (1.0)       43.7546 (31.94)         2;2      15
bench_cugraph_dgl_uniform_neighbor_sample[gpu_config=SG-dataset=rmat_25_4-with_replacement=False-fanout=10_25-batch_size=500]         94.5904 (1.13)      72.0702 (52.61)         1;1      14
bench_cugraph_dgl_uniform_neighbor_sample[gpu_config=SG-dataset=rmat_25_4-with_replacement=False-fanout=10_25-batch_size=1000]        89.1689 (1.07)       2.1179 (1.55)          1;0       3
bench_cugraph_dgl_uniform_neighbor_sample[gpu_config=SG-dataset=rmat_25_4-with_replacement=False-fanout=10_25-batch_size=2500]       141.4596 (1.69)      96.7475 (70.63)         2;2      11
bench_cugraph_dgl_uniform_neighbor_sample[gpu_config=SG-dataset=rmat_25_4-with_replacement=False-fanout=10_25-batch_size=5000]       155.7184 (1.86)     106.3896 (77.67)         2;2      10
bench_cugraph_dgl_uniform_neighbor_sample[gpu_config=SG-dataset=rmat_25_4-with_replacement=False-fanout=10_25-batch_size=10000]      188.4561 (2.25)     121.4741 (88.68)         2;2       9
bench_cugraph_dgl_uniform_neighbor_sample[gpu_config=SG-dataset=rmat_25_4-with_replacement=False-fanout=10_25-batch_size=20000]      201.9263 (2.42)     119.8722 (87.51)         2;0       8
bench_cugraph_dgl_uniform_neighbor_sample[gpu_config=SG-dataset=rmat_25_4-with_replacement=False-fanout=10_25-batch_size=30000]      222.6969 (2.66)     129.3296 (94.41)         2;0       7
bench_cugraph_dgl_uniform_neighbor_sample[gpu_config=SG-dataset=rmat_25_4-with_replacement=False-fanout=10_25-batch_size=40000]      229.5216 (2.75)     118.9508 (86.84)         2;0       7
bench_cugraph_dgl_uniform_neighbor_sample[gpu_config=SG-dataset=rmat_25_4-with_replacement=False-fanout=10_25-batch_size=50000]      153.5004 (1.84)       1.3698 (1.0)           1;0       3
bench_cugraph_dgl_uniform_neighbor_sample[gpu_config=SG-dataset=rmat_25_4-with_replacement=False-fanout=10_25-batch_size=60000]      225.2665 (2.69)     111.0862 (81.09)         2;0       7
bench_cugraph_dgl_uniform_neighbor_sample[gpu_config=SG-dataset=rmat_25_4-with_replacement=False-fanout=10_25-batch_size=70000]      246.0592 (2.94)     125.3636 (91.52)         2;0       7
bench_cugraph_dgl_uniform_neighbor_sample[gpu_config=SG-dataset=rmat_25_4-with_replacement=False-fanout=10_25-batch_size=80000]      179.6898 (2.15)       6.1373 (4.48)          1;0       3
bench_cugraph_dgl_uniform_neighbor_sample[gpu_config=SG-dataset=rmat_25_4-with_replacement=False-fanout=10_25-batch_size=90000]      279.9968 (3.35)     135.1328 (98.65)         2;0       6
bench_cugraph_dgl_uniform_neighbor_sample[gpu_config=SG-dataset=rmat_25_4-with_replacement=False-fanout=10_25-batch_size=100000]     212.5346 (2.54)      35.1669 (25.67)         1;0       3
-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

@codecov-commenter
Copy link

Codecov Report

❗ No coverage uploaded for pull request base (branch-23.02@c745570). Click here to learn what that means.
Patch has no changes to coverable lines.

Additional details and impacted files
@@               Coverage Diff               @@
##             branch-23.02    #3078   +/-   ##
===============================================
  Coverage                ?   66.99%           
===============================================
  Files                   ?       12           
  Lines                   ?      712           
  Branches                ?        0           
===============================================
  Hits                    ?      477           
  Misses                  ?      235           
  Partials                ?        0           

Help us with your feedback. Take ten seconds to tell us how you rate us. Have a feature suggestion? Share it here.

☔ View full report at Codecov.
📢 Do you have feedback about the report comment? Let us know in this issue.

@VibhuJawa VibhuJawa closed this Dec 20, 2022
rapids-bot bot pushed a commit that referenced this pull request Jan 5, 2023
This PR is a replacement of #3078 . Needed a new PR to avoid a bunch of conflict resolution. 

Example command: 
```python
DASK_NUM_WORKERS=8 pytest bench_cugraph_dgl_uniform_neighbor_sample.py -k "MG and fanout_10_25 and rmat_26_8" gi--benchmark-save='8_rmat_26_8.json'
```

Authors:
  - Vibhu Jawa (https://github.com/VibhuJawa)
  - Rick Ratzel (https://github.com/rlratzel)

Approvers:
  - Rick Ratzel (https://github.com/rlratzel)
  - AJ Schmidt (https://github.com/ajschmidt8)

URL: #3092
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.

2 participants