Uploaded image for project: 'OpenShift Workloads'
  1. OpenShift Workloads
  2. WRKLDS-1500

Utilization based Profile for descheduler for load aware descheduling

XMLWordPrintable

    • Icon: Epic Epic
    • Resolution: Unresolved
    • Icon: Critical Critical
    • None
    • None
    • descheduler-utilization-profile
    • Improvement
    • False
    • None
    • False
    • Not Selected
    • To Do
    • OCPSTRAT-1675 - Descheduler custom profile based on utilization of nodes- Dev P (December 2024)
    • OCPSTRAT-1675Descheduler custom profile based on utilization of nodes- Dev P (December 2024)

      Epic Goal*

      What is our purpose in implementing this?  What new capability will be available to customers?

      1. A new profile to trigger evictions based on node utilization / load specified via metrics
      2. A new eviction logic to determine pod eviction
       
      Why is this important? (mandatory)

      Load aware descheduling is expected to improve the performance of workloads because the aggregate load of all workloads on a cluster is spread more evenly accross multiple nodes, instead of hogging on some "hot" nodes.
       
      Scenarios (mandatory) 

      Provide details for user scenarios including actions to be performed, platform specifications, and user personas.  

      1.  TBD

       
      Dependencies (internal and external) (mandatory)

      This epic relies on the descheduler only

      Contributing Teams(and contacts) (mandatory) 

      Our expectation is that teams would modify the list below to fit the epic. Some epics may not need all the default groups but what is included here should accurately reflect who will be involved in delivering the epic.

      • Development - Virtualization CE
      • Documentation -
      • QE - 
      • PX - 
      • Others -

      Acceptance Criteria (optional)

      Provide some (testable) examples of how we will know if we have achieved the epic goal.  

      TBD

      Drawbacks or Risk (optional)

      Reasons we should consider NOT doing this such as: limited audience for the feature, feature will be superseded by other work that is planned, resulting feature will introduce substantial administrative complexity or user confusion, etc.

      • Load aware descheduling might not be enough, we might need to taint busy nodes in order to avoid workloads to land on them. However, this is a follow up.

      Done - Checklist (mandatory)

      The following points apply to all epics and are what the OpenShift team believes are the minimum set of criteria that epics should meet for us to consider them potentially shippable. We request that epic owners modify this list to reflect the work to be completed in order to produce something that is potentially shippable.

      • CI Testing -  Basic e2e automationTests are merged and completing successfully
      • Documentation - Content development is complete.
      • QE - Test scenarios are written and executed successfully.
      • Technical Enablement - Slides are complete (if requested by PLM)
      • Engineering Stories Merged
      • All associated work items with the Epic are closed
      • Epic status should be "Release Pending" 

            Unassigned Unassigned
            fdeutsch@redhat.com Fabian Deutsch
            Votes:
            0 Vote for this issue
            Watchers:
            1 Start watching this issue

              Created:
              Updated: