Fix Hyperdisk Resize That Requires Iops/Throughput Adjustment

[sunnylovestiramisu]

What type of PR is this?
/kind bug

What this PR does / why we need it:
Currently resizing for Hyperdisk is failing for cases that require IOPS or throughput getting updated in the same resize call.

Which issue(s) this PR fixes:
A hyperdisk-balanced with size 4Gi needs exactly 2000 IOPS, and 5Gi needs exactly 2500 IOPS. Disks 6Gi and more need a minimum of 3000 IOPS. When resizing a hyperdisk, there is a need to update IOPS as well.

For hyperdisk-extreme the min IOPS value is 2 IOPS per GiB, if a 64 GiB hyperdisk-extreme disk with 128 IOPS is resized to 65 GiB, the IOPS also need to be updated to 130.

For hyperdisk-ml it is more complicated, min IOPS is 16 IOPS per MBps of provisioned throughput, min throughput is the greater of 0.12 MBps per GiB and 400 MBps per disk. A 3334GiB hyperdisk-ml with 400 MBps throughput and initial 6400 IOPS resize to larger capacity will also need to update both throughput and IOPS.

Fixes #

Special notes for your reviewer:
Tested end to end in GCE clusters.

Type	From Size to Size	Notes
hyperdisk-balanced	4 Gib to 5Gi	n4-standard-4 IOPS: 2000 -> 2500 Throughput: 140MB/s(no change)
hyperdisk-balanced	5Gi to 20Gi	n4-standard-4 IOPS: 2500->3000 Throughput: 145MB/s(no change)
hyperdisk-balanced-high-availability	5Gi to 6Gi	c3-standard-4 IOPS: 2500->3000 Throughput: 145MB/s(no change)
hyperdisk-extreme	64Gi to 70Gi	IOPS: 128 to 140
hyperdisk-throughput	2Ti to 3Ti	c3-standard-4 Throughput: 20MB/s to 30MB/s
hyperdisk-ml	3334Gi to 3500Gi	c3-standard-4 Throughput: 400MB/s to 420MB/s

Does this PR introduce a user-facing change?:

Fix Hyperdisk Resize That Requires Iops/Throughput Adjustment

[sunnylovestiramisu]

/test pull-gcp-compute-persistent-disk-csi-driver-e2e

[sunnylovestiramisu]

/assign @msau42

[sunnylovestiramisu]

/test pull-gcp-compute-persistent-disk-csi-driver-verify

[sunnylovestiramisu]

/test pull-gcp-compute-persistent-disk-csi-driver-e2e

[mattcary]

This should be easy to add an e2e test for the presubmit?

Let's submit this now though and add e2e later (which I think would be preferable to an internal GKE test).

The e2e test could just be for the simple HdB case, testing thoughput and such gets harder with machine compatibility.

/lgtm
/approve

[k8s-ci-robot]

[APPROVALNOTIFIER] This PR is APPROVED

This pull-request has been approved by: mattcary, sunnylovestiramisu

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 [mattcary,sunnylovestiramisu]

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

[sunnylovestiramisu]

/cherry-pick release-1.17

[k8s-infra-cherrypick-robot]

@sunnylovestiramisu: #2054 failed to apply on top of branch "release-1.17":

Applying: Fix Hyperdisk Resize That Requires Iops/Throughput Adjustment
Using index info to reconstruct a base tree...
M	pkg/common/utils.go
Falling back to patching base and 3-way merge...
Auto-merging pkg/common/utils.go
CONFLICT (content): Merge conflict in pkg/common/utils.go
error: Failed to merge in the changes.
hint: Use 'git am --show-current-patch=diff' to see the failed patch
hint: When you have resolved this problem, run "git am --continue".
hint: If you prefer to skip this patch, run "git am --skip" instead.
hint: To restore the original branch and stop patching, run "git am --abort".
hint: Disable this message with "git config advice.mergeConflict false"
Patch failed at 0001 Fix Hyperdisk Resize That Requires Iops/Throughput Adjustment

In response to this:

/cherry-pick release-1.17

Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes-sigs/prow repository.