Skip to content

Conversation

@keighrim
Copy link
Member

fixes #243
also updates to the latest mmif-python

* now all gunicorn workers are discarded after processing 1 request
    * this will prevent VRAM leak caused by zombie workers holding VRAM doing nothing
    * but in exchange of some loading time at worker initialization (e.g. model loading)
    * can be overridden by setting `max_requests` passed to `serve_production()` in app.py
* number of gunicorn workers are set reasonably low when app declares GPU usage
* app metadata now has two new fields can be used to declare GPU usage
* for GPU apps, first-time VRAM usage for each parameter combination is now recorded in local cache directory see `$XDG_CACHE_HOME/clams`
@clams-bot clams-bot added this to infra Nov 22, 2025
@github-project-automation github-project-automation bot moved this to Todo in infra Nov 22, 2025
@codecov
Copy link

codecov bot commented Nov 22, 2025

Codecov Report

❌ Patch coverage is 19.58042% with 115 lines in your changes missing coverage. Please review.
⚠️ Please upload report for BASE (develop@69bdca9). Learn more about missing BASE report.
⚠️ Report is 4 commits behind head on develop.

Files with missing lines Patch % Lines
clams/app/__init__.py 16.50% 86 Missing ⚠️
clams/restify/__init__.py 0.00% 29 Missing ⚠️
Additional details and impacted files
@@            Coverage Diff             @@
##             develop     #266   +/-   ##
==========================================
  Coverage           ?   61.33%           
==========================================
  Files              ?        6           
  Lines              ?      812           
  Branches           ?        0           
==========================================
  Hits               ?      498           
  Misses             ?      314           
  Partials           ?        0           
Flag Coverage Δ
unittests 61.33% <19.58%> (?)

Flags with carried forward coverage won't be shown. Click here to find out more.

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:
  • ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

@keighrim keighrim merged commit 9dcad01 into develop Nov 22, 2025
8 checks passed
@keighrim keighrim deleted the 243-fix-cuda-mem-leak branch November 22, 2025 13:52
@github-project-automation github-project-automation bot moved this from Todo to Done in infra Nov 22, 2025
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

Archived in project

Development

Successfully merging this pull request may close these issues.

gunicorn, torch, and cuda

2 participants