Serving impovements

[andrewballantyne]

Closes: #910

Description

Misc improvements to Model Serving -- see ticket for the full list.

How Has This Been Tested?

You'll need multiple projects and a few models within' to test some of the effort on the global model serving page.

Merge criteria:

• The commits are squashed in a cohesive manner and have meaningful messages.
• Testing instructions have been added in the PR body (for PRs involving changes that are not immediately obvious).
• The developer has manually tested the changes and verified that the changes work

[Gkrumbach07]

Allow model server replicas to scale to 0

odh-dashboard/frontend/src/pages/modelServing/screens/projects/ServingRuntimeModal/ManageServingRuntimeModal.tsx

Line 68 in 74ec5b1

createData.numReplicas > 0 &&

Configure becomes disabled even when a valid replica of 0 is set. Set to createData.numReplicas >= 0

[Gkrumbach07]

(2) allow Request and Limit to be the same value

Same issue as above with validation

odh-dashboard/frontend/src/pages/modelServing/screens/projects/ServingRuntimeModal/ManageServingRuntimeModal.tsx

Lines 73 to 76 in 74ec5b1

	parseInt(createData.modelSize.resources.limits.cpu) >
	parseInt(createData.modelSize.resources.requests.cpu) &&
	parseInt(createData.modelSize.resources.limits.memory) >
	parseInt(createData.modelSize.resources.requests.memory);

Also, is it an issue that a user can request a limit of 1 milicore but request a whole core (same with memory units)?
Maybe there should be an additional validation for differing units. Possibly also an error message as this could be a not obvious issue to a user

[andrewballantyne]

(2) allow Request and Limit to be the same value

Ah good call on the units -- and wow, I totally missed no2 when I was going through. Thanks!

[Gkrumbach07]

/lgtm

[lucferbux]

/lgtm
Code seems great, tested visual changes and everything seems to be checked, played around with the logic of the limits and requests and they behave as expected

[lucferbux]

/approve

[andrewballantyne]

Trying to wait on #935 as we are going to conflict and I have another PR that also conflicts on 935 -- so I agreed with Chris he can merge first (assuming he is able to clean it up in a timely manner)

[lucferbux]

Ok perfect!

[andrewballantyne]

Looks to be up to date now with the main branch after gpu addition to the model serving

[lucferbux]

Testing it now

[lucferbux]

/lgtm

[openshift-ci]

[APPROVALNOTIFIER] This PR is APPROVED

Approval requirements bypassed by manually added approval.

This pull-request has been approved by: Gkrumbach07, lucferbux

The full list of commands accepted by this bot can be found here.

The pull request process is described here

Needs approval from an approver in each of these files:

• OWNERS

Approvers can indicate their approval by writing /approve in a comment
Approvers can cancel approval by writing /approve cancel in a comment