2020-12-16 Meeting Notes

Date

16-Dec-2020

Attendees

Discussion items

TimeItemWhoNotes

Review of Performance Testing Environment recommendationTC

Martin presents this page: Transferring Performance Testing to FOLIO Teams

Discussion:

  • How to handle new teams? Same way - have them identify an individual and run
  • How long for an individual to get up to speed - need 2 test runs - about 2 weeks
  • Question on the mechanics: Since the data is collected in Influx DBs, and teams will be bringing up and tearing down environments will test data be maintained? Answer - yes, the data is persisted via an EBS volume.
  • Costs for the EBS container are included in the proposal
  • Are we essentially talking about the designated person on a team doing the tests (rather than the whole team)? No - the idea is that the person who are designated eventually train others in their team so that it's not just a single person.
  • What happens when a team finds that the source of a performance issues actually lies in a different team's code. PTF will still be around to consult, but hopefully all teams are able to use these tools
  • The "holy grail" is that all teams can do performance testing at any time they need it.
  • The implication of accepting this proposal is that TC is advocating adding these tests to teams' workload and setting the expectations that performance is tested and certified - and what are those expectations? Testing after every story?
  • Concern that this might be a lot of overhead to the teams
  • What if the Stakeholders said "you can spend half" ?

Next Steps

  • Agree with the proposal and seek funding
  • Hesitate to mandate specifics on the expectation for testing, but rather allow teams to take advantage of it for a quarter or two
  • Once we have experience with adoption and benefits we can consider more specific recommendations/expectations for teams.



Definition of Reference/Default Data Types Taskforce members

Previous update on the weekly discussions on this topic:

Summary: Team is converging - key notion of separating "Reference data" into 2 groups:

    • Operational Data
    • Default Data

Making definitions and recommendations as to how to proceed. 

Updates?

  • General agreement on workflow diagram and concepts/labels, but still discussing a few steps and final decisions.
  • Will be listing the artifacts and examples on how "conflicts" and "customized" data work out
  • Want to validate the workflow and plan with SysOps before bringing it back to TC
  • Not going to be meeting for the next few weeks (Holidays)
  • Hope to be able to present to TC in late January