DEVPROD-19870: Downsize instances where possible

[anjani-bhat]

Ticket: DEVPROD-19870
This PR is for cutting instance sizes to use high memory variants. This should result in ~ 15k savings in a year.

Proof of Work

Here are the successful patches for the variants which were downsized:
e2e_om60_kind_ubi
patch
This results in the variant being 3-4% slower (considering total duration not wall time) on an average.
The longest task went from
47m 58s to 48m 53s, so the wall clock time change would be ~2%. The cost savings would be ~30-32%.

e2e_om70_kind_ubi
patch
This results in the variant being 3-4% slower (considering total duration not wall time) on an average.
The longest task went from
48m 12s to 47m 8s, so the wall clock time shouldn't change by much. The cost savings would be ~32%.

e2e_om80_kind_ubi
patch
This results in the variant being 3-4% slower (considering total duration not wall time) on an average.
The longest task went from
45m 2s to 45m 51s, so the wall clock time should change be 1-2%. The cost savings would be ~30-32%.

Checklist

• Have you linked a jira ticket and/or is the ticket in the title?
• Have you checked whether your jira ticket required DOCSP changes?
• Have you added changelog file?
  • use skip-changelog label if not needed
  • refer to Changelog files and Release Notes section in CONTRIBUTING.md for more details

[anjani-bhat]

@anandsyncs: I don't have write access to this repo, so could you add the skip-changelog label for this PR? Also let me know if anything else needs to be done for this PR. Thank you !

[nammn]

Thanks!
Have you've checked the same for e2e_static_om.*_kind_ubi variants?

And have you've considered verifying via multiple patches?

[anjani-bhat]

Have you've checked the same for e2e_static_om.*_kind_ubi variants?

Thanks for calling this out. Downsized e2e_static_om60_kind_ubi as well. The other e2e_static_om.*_kind_ubi variants are already on the smaller variants.

And have you've considered verifying via multiple patches?

Yes, here are the patches:

• https://spruce.mongodb.com/version/68a39abace035a00085936d9
• https://spruce.mongodb.com/version/689e3aab9ad11a0007d88ee7
• https://spruce.mongodb.com/version/68a88bf0060b5b0008f6294a [the failed task seems to be failing a lot in the waterfall as well for e2e_om80_kind_ubi with the same error (see here for mainline error and here for patch error)]
• additional patch for e2e_static_om60_kind_ubi: https://spruce.mongodb.com/version/68a88b182b0df90007152e10

[nammn]

thanks a lot!
But why not also update

e2e_static_om70_kind_ubi and e2e_static_om80_kind_ubi?

Anyway, please update them as well if patches show the same effect

[anjani-bhat]

e2e_static_om70_kind_ubi and e2e_static_om80_kind_ubi?

They are already updated
https://github.com/mongodb/mongodb-kubernetes/blob/master/.evergreen.yml#L1362
https://github.com/mongodb/mongodb-kubernetes/blob/master/.evergreen.yml#L1385

[anandsyncs]

evergreen retry

[anandsyncs]

@anjani-bhat You probably have collaborator access, can you please raise this PR directly from this repo?

[anjani-bhat]

@anjani-bhat You probably have collaborator access, can you please raise this PR directly from this repo?

@anandsyncs I don't have permissions to push to mongodb/mongodb-kubernetes. I can reach out to @mircea-cosbuc since he was able to merge the last PR.