Skip to content
GitLab
Projects Groups Topics Snippets
  • /
  • Help
    • Help
    • Support
    • Community forum
    • Submit feedback
  • Sign in
  • Moore Moore
  • Project information
    • Project information
    • Activity
    • Labels
    • Members
  • Repository
    • Repository
    • Files
    • Commits
    • Branches
    • Tags
    • Contributor statistics
    • Graph
    • Compare revisions
    • Locked files
  • Issues 259
    • Issues 259
    • List
    • Boards
    • Service Desk
    • Milestones
    • Iterations
    • Requirements
  • Jira
    • Jira
  • Merge requests 98
    • Merge requests 98
  • CI/CD
    • CI/CD
    • Pipelines
    • Jobs
    • Artifacts
    • Schedules
    • Test cases
  • Deployments
    • Deployments
    • Environments
    • Releases
  • Packages and registries
    • Packages and registries
    • Model experiments
  • Monitor
    • Monitor
    • Incidents
  • Analytics
    • Analytics
    • CI/CD
    • Code review
    • Insights
    • Issue
    • Repository
  • Snippets
    • Snippets
  • Activity
  • Graph
  • Create a new issue
  • Jobs
  • Commits
  • Issue Boards
Collapse sidebar

Admin message

GitLab Runners cleaning campaign ongoing (ETA 2nd October 2023). Check https://cern.ch/otg0078219 for further information.

  • LHCbLHCb
  • MooreMoore
  • Issues
  • #507
Closed
Open
Issue created Jan 09, 2023 by Mika Anton Vesterinen@mvesteriDeveloper

Hlt2 bandwidth division tools

We already have the nice UpgradeRateTest dashboard developed by @shunan @abertoli et al.

Following a discussion in the RTA WP3 meeting today it seems that the main missing feature is the ability to report correlations and exclusive rates. This is not possible by simply parsing the log file. It requires writing out a tuple of decisions with one column per line and one row per event, like HltEfficiencyChecker does (this is also what the Run-2 rate tests did). The analysis on the tuple to produce content for the webpages is then very fast/efficient.

One technical issue raised by @shunan is that the tests are currently run multithreaded, which probably rules out any tupling based on {MC}DecayTreeTuple (what HltEfficiencyChecker currently does)

The work by @enoomen slides is also relevant.

@poluekt @nskidmor @shunan @abertoli @msaur @sstahl @enoomen @rmatev @rjhunter

cc @cmarinbe @decianm

Assignee
Assign to
Time tracking