Add weekly tests for memory growth #3101
Merged
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
krishung5
changed the title
CoderHam
suggested changes
Jul 9, 2021
CoderHam
suggested changes
Jul 9, 2021
CoderHam
suggested changes
Jul 9, 2021
CoderHam
previously approved these changes
Jul 9, 2021
krishung5
changed the title
GuanLuo
reviewed
Jul 12, 2021
deadeyegoodwin
suggested changes
Jul 12, 2021
Dockerfile.QA
Outdated
| @@ -191,6 +191,12 @@ RUN if [ -d qa/L0_model_control_stress ]; then \ | |||
| cp -r qa/L0_model_control_stress/. qa/L0_model_control_stress_valgrind_massif; \ | |||
| fi | |||
|
|
|||
| # Create L0_client_memory_growth_weekly and L0_memory_growth_weekly for weekly test | |||
There was a problem hiding this comment.
You shouldn't need to copy or rename the tests. We are just going to run the existing tests weekly (instead of nightly). We may make changes to the existing tests but we can to that in the existing code.
| @@ -35,7 +35,11 @@ | |||
|
|
|||
| if __name__ == '__main__': | |||
| today = date.today().strftime("%Y-%m-%d") | |||
| subject = "Triton Client Memory Growth Summary: " + today | |||
| # Set the subject for weekly and nightly tests | |||
There was a problem hiding this comment.
You can keep this, but for now we will just run this test weekly.
| @@ -35,7 +35,11 @@ | |||
|
|
|||
| if __name__ == '__main__': | |||
| today = date.today().strftime("%Y-%m-%d") | |||
| subject = "Triton Server Memory Growth Summary: " + today | |||
| # Set the subject for weekly and nightly tests | |||
| if (sys.argv[1] == "weekly"): | |||
There was a problem hiding this comment.
You can keep this, but for now we will just run this test weekly.
There was a problem hiding this comment.
Since the argument is now a string, you don't need the conditional, instead, you can just set the subject variable based on the argument subject = "Triton Server Memory Growth " + sys.argv[1] + " Summary: " + today
GuanLuo
previously approved these changes
Jul 13, 2021
deadeyegoodwin
previously approved these changes
Jul 13, 2021
CoderHam
suggested changes
Jul 13, 2021
CoderHam
previously approved these changes
Jul 13, 2021
CoderHam
suggested changes
Jul 14, 2021
qa/L0_memory_growth/test.sh
Outdated
| RET=1 | ||
| fi | ||
| set -e | ||
| # The busy op model causes issues when running the CI. |
There was a problem hiding this comment.
Call out PTX here so we can quickly re-enable tests that were disabled due to PTX fsilures
Also make this a TODO
# TODO Re-enable after PTX issues are resolved.
1bcf6e9
to
8e5783b
Compare
CoderHam
previously approved these changes
Jul 14, 2021
CoderHam
suggested changes
Jul 15, 2021
| @@ -37,10 +37,12 @@ | |||
| today = date.today().strftime("%Y-%m-%d") | |||
| subject = "Triton Client Memory Growth " + sys.argv[1] + " Summary: " + today | |||
| memory_graphs = glob.glob("client_memory_growth*.log") | |||
| html_content = "<html><head></head><body><pre style=\"font-size:11pt;font-family:Consolas;\">" | |||
| write_up = "<p>This test is run for both HTTP and GRPC protocols using C++ and Python test scripts. The thresholds of memory growth are set to 10MB and 1MB for C++ and Python tests individually as the max-allowed difference between mean and maximum memory usage.</p>" | |||
There was a problem hiding this comment.
The threshold is not 10MB, the max difference between mean and max memory usage is 10MB
| for mem_graph in sorted(memory_graphs): | ||
| html_content += "\n" + mem_graph + "\n" | ||
| with open(mem_graph, "r") as f: | ||
| html_content += f.read() + "\n" | ||
| html_content += "</pre></body></html>" | ||
| nightly_email_helper.send(subject, html_content, is_html=True) | ||
| nightly_email_helper.send(subject, html_content, is_html=True) |
There was a problem hiding this comment.
Add back empty line at end
| for mem_graph in sorted(memory_graphs): | ||
| memory_graphs_resnet = glob.glob("memory_growth_resnet*.log") | ||
| memory_graphs_busyop = glob.glob("memory_growth_busyop.log") | ||
| write_up = "<p>This test uses perf_analyzer as clients running on 4 different models. The threshold of memory growth is set to 150MB as the max allowed difference between mean and maximum memory usage.</p>" |
CoderHam
suggested changes
Jul 15, 2021
Dockerfile.QA
Outdated
| @@ -1,4 +1,4 @@ | |||
| # Copyright (c) 2018-2021, NVIDIA CORPORATION. All rights reserved. | |||
| # Copyright (c) 2018-2021, NVIDIA CORPORATION & AFFILIATES. All rights reserved. | |||
There was a problem hiding this comment.
Remove changes for copyright in files that were not modified.
CoderHam
previously approved these changes
Jul 15, 2021
added 25 commits
July 16, 2021 10:13
krishung5
force-pushed
the
krish-memory-weekly-tests
branch
from
edfbdcc
to
ab05849
Compare
CoderHam
approved these changes
Jul 16, 2021
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.
No description provided.