test: TC for Metric P0 nv_load_time per model #7697
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
indrajit96
changed the title
19 tasks
rmccorm4
reviewed
Oct 14, 2024
rmccorm4
reviewed
Oct 14, 2024
rmccorm4
reviewed
Oct 14, 2024
kthui
reviewed
Oct 19, 2024
Comment on lines
+147
to
+161
| set +e | ||
| CLIENT_PY="./general_metrics_test.py" | ||
| CLIENT_LOG="general_metrics_test_client.log" | ||
| SERVER_LOG="general_metrics_test_server.log" | ||
| SERVER_ARGS="$BASE_SERVER_ARGS --model-control-mode=explicit --log-verbose=1" | ||
| run_and_check_server | ||
| MODEL_NAME='libtorch_float32_float32_float32' | ||
| code=`curl -s -w %{http_code} -X POST ${TRITONSERVER_IPADDR}:8000/v2/repository/models/${MODEL_NAME}/load` | ||
| # Test 2 for explicit mode LOAD | ||
| python3 -m pytest --junitxml="general_metrics_test.test_metrics_load_time_explicit_load.report.xml" $CLIENT_PY::TestGeneralMetrics::test_metrics_load_time_explicit_load >> $CLIENT_LOG 2>&1 | ||
|
|
||
| code=`curl -s -w %{http_code} -X POST ${TRITONSERVER_IPADDR}:8000/v2/repository/models/${MODEL_NAME}/unload` | ||
| # Test 3 for explicit mode UNLOAD | ||
| python3 -m pytest --junitxml="general_metrics_test.test_metrics_load_time_explicit_unload.report.xml" $CLIENT_PY::TestGeneralMetrics::test_metrics_load_time_explicit_unload >> $CLIENT_LOG 2>&1 | ||
| kill_server |
There was a problem hiding this comment.
For test 2 and 3, I think it would be more helpful by testing the following model load/unload sequence against metrics:
- Start the server without loading any model.
- Check the metrics is empty.
- Load a few models (and have one model with two versions that loads at a different speed).
- Check the metrics is correct.
- Call the load API again (without changing the model repository).
- Check the metrics is unchanged.
- Load a new model.
- Check the metrics is updated correctly.
- Unload models.
- Check the metrics is unchanged.
…-server/server into ibhosale_metrics_google
![github-advanced-security[bot]](https://avatars.githubusercontent.com/in/57789?s=60&v=4){kind=small}
| def get_model_load_times(): | ||
| r = requests.get(f"http://{_tritonserver_ipaddr}:8002/metrics") | ||
| r.raise_for_status() | ||
| pattern = re.compile(rf'{MODEL_LOAD_TIME}"(.*?)".*?\ (\d+\.\d+)') |
Check notice
Code scanning / CodeQL
Unused local variable Note
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
What does the PR do?
Test Case of model load time metrics
Checklist
<commit_type>: <Title>Commit Type:
Check the conventional commit type
box here and add the label to the github PR.
Related PRs:
Core : triton-inference-server/core#397
Where should the reviewer start?
qa/L0_metrics/general_metrics_test.py
Test plan:
Added tests for
Background
Improve metrics in Triton