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

[Data] [CI] Add batch inference benchmarks #37283

Merged
merged 15 commits into from
Jul 25, 2023

Conversation

amogkam
Copy link
Contributor

@amogkam amogkam commented Jul 11, 2023

Adds the following benchmarks as release tests

  1. Image classification on 10 GB raw image data, 1 GPU
  2. Image classification on 10 GB parquet data, 1 GPU
  3. Image classification on 300 GB raw image data, 16 GPUs
  4. Image classification on 300 GB parquet data, 16 GPUs
  5. Image classification on 10 TB parquet data, 40 GPUs, weekly frequency

Closes #37345
Closes #28174
Closes #31796

Why are these changes needed?

Related issue number

Checks

  • I've signed off every commit(by using the -s flag, i.e., git commit -s) in this PR.
  • I've run scripts/format.sh to lint the changes in this PR.
  • I've included any doc changes needed for https://docs.ray.io/en/master/.
    • I've added any new APIs to the API Reference. For example, if I added a
      method in Tune, I've added it in doc/source/tune/api/ under the
      corresponding .rst file.
  • I've made sure the tests are passing. Note that there might be a few flaky tests, see the recent failures at https://flakey-tests.ray.io/
  • Testing Strategy
    • Unit tests
    • Release tests
    • This PR is not tested :(

Signed-off-by: amogkam <[email protected]>
Signed-off-by: amogkam <[email protected]>
Signed-off-by: amogkam <[email protected]>
Signed-off-by: amogkam <[email protected]>
Copy link
Contributor

@c21 c21 left a comment

Choose a reason for hiding this comment

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

LG w/ some minor comments.


# 10 GB image classification raw images with 1 GPU.
# 1 g4dn.4xlarge
- name: air_benchmark_torch_batch_prediction_1_gpu_10gb_parquet
Copy link
Contributor

Choose a reason for hiding this comment

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

do we still want to keep (10GB, 1GPU), given we already have (300GB, 16GPU) to run nightly?

Copy link
Contributor Author

Choose a reason for hiding this comment

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

I think it can help...it's a pretty cheap job cost wise and can help isolate whether regressions are occurring in the single node/single GPU case, or only in the multi node case. But I am open to removing it.

release/BUILD Outdated Show resolved Hide resolved
Signed-off-by: amogkam <[email protected]>
Signed-off-by: amogkam <[email protected]>
Signed-off-by: amogkam <[email protected]>
Signed-off-by: amogkam <[email protected]>
@c21
Copy link
Contributor

c21 commented Jul 21, 2023

@amogkam - is it good to merge?

@amogkam
Copy link
Contributor Author

amogkam commented Jul 22, 2023

@c21, no, let me kick off the release test

Signed-off-by: amogkam <[email protected]>
@amogkam
Copy link
Contributor Author

amogkam commented Jul 24, 2023

10 tb test passing with 11580.81978035109 images/second

https://buildkite.com/ray-project/release-tests-pr/builds/46310

Signed-off-by: amogkam <[email protected]>
@amogkam amogkam merged commit e1a97c3 into ray-project:master Jul 25, 2023
2 checks passed
@amogkam amogkam deleted the batch-inference-release-tests branch July 25, 2023 00:58
NripeshN pushed a commit to NripeshN/ray that referenced this pull request Aug 15, 2023
Adds the following benchmarks as release tests

Image classification on 10 GB raw image data, 1 GPU
Image classification on 10 GB parquet data, 1 GPU
Image classification on 300 GB raw image data, 16 GPUs
Image classification on 300 GB parquet data, 16 GPUs
Image classification on 10 TB parquet data, 40 GPUs, weekly frequency

---------

Signed-off-by: amogkam <[email protected]>
Signed-off-by: NripeshN <[email protected]>
harborn pushed a commit to harborn/ray that referenced this pull request Aug 17, 2023
Adds the following benchmarks as release tests

Image classification on 10 GB raw image data, 1 GPU
Image classification on 10 GB parquet data, 1 GPU
Image classification on 300 GB raw image data, 16 GPUs
Image classification on 300 GB parquet data, 16 GPUs
Image classification on 10 TB parquet data, 40 GPUs, weekly frequency

---------

Signed-off-by: amogkam <[email protected]>
Signed-off-by: harborn <[email protected]>
harborn pushed a commit to harborn/ray that referenced this pull request Aug 17, 2023
Adds the following benchmarks as release tests

Image classification on 10 GB raw image data, 1 GPU
Image classification on 10 GB parquet data, 1 GPU
Image classification on 300 GB raw image data, 16 GPUs
Image classification on 300 GB parquet data, 16 GPUs
Image classification on 10 TB parquet data, 40 GPUs, weekly frequency

---------

Signed-off-by: amogkam <[email protected]>
arvind-chandra pushed a commit to lmco/ray that referenced this pull request Aug 31, 2023
Adds the following benchmarks as release tests

Image classification on 10 GB raw image data, 1 GPU
Image classification on 10 GB parquet data, 1 GPU
Image classification on 300 GB raw image data, 16 GPUs
Image classification on 300 GB parquet data, 16 GPUs
Image classification on 10 TB parquet data, 40 GPUs, weekly frequency

---------

Signed-off-by: amogkam <[email protected]>
Signed-off-by: e428265 <[email protected]>
vymao pushed a commit to vymao/ray that referenced this pull request Oct 11, 2023
Adds the following benchmarks as release tests

Image classification on 10 GB raw image data, 1 GPU
Image classification on 10 GB parquet data, 1 GPU
Image classification on 300 GB raw image data, 16 GPUs
Image classification on 300 GB parquet data, 16 GPUs
Image classification on 10 TB parquet data, 40 GPUs, weekly frequency

---------

Signed-off-by: amogkam <[email protected]>
Signed-off-by: Victor <[email protected]>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
4 participants