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.

Sign up for GitHub

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

Jump to bottom

Improve NCU Reader #60

Merged
merged 12 commits into from
Apr 25, 2024
Merged

Improve NCU Reader #60

merged 12 commits into from
Apr 25, 2024

Conversation

michaelmckinsey1
Copy link
Collaborator

@michaelmckinsey1 michaelmckinsey1 commented Jul 10, 2023 •
edited
Loading

Summary

Improve the NCU Reader.

  1. Leverage nvtx push_pop_ranges call trace and Hatchet Query Language to improve node-mapping logic accuracy.
  2. Add aggregation of kernel reps instead of discarding duplicate kernel runs. 4 different aggregation options available, mean, max, min, sum, programmatically determined from ncu's rollup_operation metric function.
  3. Improve interface by making NCU a thicket class function.
  4. Refactor NCU into a class object.
  5. Add tqdm to main loop in reader to provide some clarity to the user.

@michaelmckinsey1 michaelmckinsey1 added area-external Issues and PRs related to external libraries used by Thicket priority-normal Normal priority issues and PRs status-ready-for-review This PR is ready to be reviewed by assigned reviewers type-feature Requests for new features or PRs which implement new features labels Jul 11, 2023
@slabasan slabasan modified the milestones: 2023.3.0, 2023.3.1 Aug 25, 2023
@michaelmckinsey1 michaelmckinsey1 removed this from the 2023.3.1 milestone Apr 15, 2024
@slabasan slabasan merged commit bf7cbf4 into LLNL:develop Apr 25, 2024
4 checks passed
Yejashi pushed a commit to TauferLab/thicket that referenced this pull request Apr 28, 2024
@michaelmckinsey1 @Yejashi
Improve NCU Reader (LLNL#60)
f4c5c6b 
* Refactor NCU into a reader class and improve logic

* Remove statefulness of NCUReader

* fix query

* Add rollup operation aggregation

* Optimize rollup dictionary creation

* Optimize metric value assignment

* Add progress bar description

* Replace depreciated code
@slabasan slabasan added this to the 2024.2.0 milestone Sep 6, 2024
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers

@slabasan slabasan slabasan approved these changes

Assignees
No one assigned
Labels
area-external Issues and PRs related to external libraries used by Thicket priority-normal Normal priority issues and PRs status-ready-for-review This PR is ready to be reviewed by assigned reviewers type-feature Requests for new features or PRs which implement new features
Projects
None yet
Milestone
2024.2.0
Development

Successfully merging this pull request may close these issues.
2 participants
@michaelmckinsey1 @slabasan