Skip to content

Draft: Compact the amount of API pods running per node

John Skarbekrequested to merge
jts/compact-api-pods into master

What does this MR do?

  • This is performed by reducing the amount of CPU requests we are making for the API deployment
  • Our average CPU usage is currently below 2 CPU, but we are requesting 4
  • With these values, we drop the amount of total requested CPU down to 3100m (2500 for webservice + 600 for workhorse)
  • With a lower CPU request, using the same node type, we can reduce the amount of requested CPU per pod from the currently 28%, down to roughly 18%. Doing so should allow upwards of 5 Pods per node (pending other resources deployed)

Author Check-list

Please read the Contributing document and once you do, complete the following:

  • Assign to the correct reviewer per the contributing document
  • Apply the correct metadata per the contributing document

Reviewer Check-list

  • Reviewed the diff jobs to confirm changes are as expected
  • No changes shown in the diffs not associated with this MR - This may require a rebase or further investigation
Edited by John Skarbek

Merge request reports