OCP/Telco Definition of Done
Epic Template descriptions and documentation.

Epic Goal

• The OpenShift profiles NTO have been created for RHEL 7.x hosts.  Recent tests performed on RHEL 9.x hosts showed that the default/out-of-the-box setting of the `sched_` tunables results in similar performance with using the modified values shipped by `TuneD` profiles.  Therefore, for RHEL 9.x systems most `sched_` tunables are being dropped.  See:
  • https://github.com/redhat-performance/tuned/commit/c6d6fdcc4c944df9998e0ebe75f31cc8aed452c1

• • https://bugzilla.redhat.com/show_bug.cgi?id=1957829
    There is also talk of dropping the `sched_` tunables even for RHEL 8.x hosts due to negligible performance impact.

• • RHEL QE has done testing without the sched_ tunables on RHEL9 only, and found no performance impact.

• Determine a set of relevant tunables which impact control-plane performance, and their ranges.
• Determine a set of workloads for testing control plane performance.
• Use the kruize/autotune project to automate the optimization of the set of relevant tunables for control plane performance.
• Conduct large scale testing, ideally with the collaboration by the Perf&Scale team with/without the `sched_` tunables present in the openshift-control-plane profile, compare the results and adjust the profile for future releases accordingly.
• Provide an updated documentation on the profiles used in the OpenShift docs.

Why is this important?

• The changes that worked well on RHCOS/RHEL 7.x systems might be affect performance of the current RHCOS/RHEL 8.x / 9.x systems.
• Want to keep consistency between profiles on RHEL and RHCOS for similar workloads.
• Potential to develop a system for automating review of tuned profiles going forward

Scenarios

1. ...

Acceptance Criteria

• CI - MUST be running successfully with tests automated
• Release Technical Enablement - Provide necessary release enablement details and documents.
• ...

Dependencies (internal and external)

1. ...

Previous Work (Optional):

1. RHEL 9.x testing performed by the BaseOS engineering and QE teams.

Open questions::

1. Testing strategy to evaluate the performance impact of on the RHCOS/RHEL nodes.  Ideally, a typical set (or at least a subset) of tests run by the Perf&Scale team.  We need to focus on tests results of which the performance of the control plane particularly influences.

Done Checklist

• CI - CI is running, tests are automated and merged.
• Release Enablement <link to Feature Enablement Presentation>
• DEV - Upstream code and tests merged: <link to meaningful PR or GitHub Issue>
• DEV - Upstream documentation merged: <link to meaningful PR or GitHub Issue>
• DEV - Downstream build attached to advisory: <link to errata>
• QE - Test plans in Polarion: <link or reference to Polarion>
• QE - Automated tests merged: <link or reference to automated tests>
• DOC - Downstream documentation merged: <link to meaningful PR>