Update GPU support for V100,T4,A5500. by pdion891 · Pull Request #6639 · apache/cloudstack

pdion891 · 2022-08-15T14:08:41Z

Description

Update GPU and vGPU support for Nvidia V100, T4 and A5500 cards.

Types of changes

Breaking change (fix or feature that would cause existing functionality to change)
New feature (non-breaking change which adds functionality)
Bug fix (non-breaking change which fixes an issue)
Enhancement (improves an existing feature and functionality)
Cleanup (Code refactoring and cleanup, that may add test cases)

Feature/Enhancement Scale or Bug Severity

Feature/Enhancement Scale

Major
Minor

How Has This Been Tested?

acs-robot · 2022-08-15T14:09:44Z

Found UI changes, kicking a new UI QA build
@blueorangutan ui

blueorangutan · 2022-08-15T14:10:03Z

@acs-robot a Jenkins job has been kicked to build UI QA env. I'll keep you posted as I make progress.

blueorangutan · 2022-08-15T14:15:07Z

UI build: ✔️
Live QA URL: http://qa.cloudstack.cloud:8080/client/pr/6639 (SL-JID-2149)

DaanHoogland · 2022-08-15T14:48:31Z

@blueorangutan package

blueorangutan · 2022-08-15T14:49:03Z

@DaanHoogland a Jenkins job has been kicked to build packages. It will be bundled with KVM, XenServer and VMware SystemVM templates. I'll keep you posted as I make progress.

sonarqubecloud · 2022-08-15T15:06:25Z

SonarCloud Quality Gate failed.

0 Bugs
0 Vulnerabilities
0 Security Hotspots
0 Code Smells

0.0% Coverage
0.0% Duplication

blueorangutan · 2022-08-15T15:36:00Z

Packaging result: ✔️ el7 ✔️ el8 ✔️ debian ✔️ suse15. SL-JID 3990

DaanHoogland · 2022-08-15T16:00:47Z

@pdion891 could you advice on how to test this, or report on your own testing?

codecov · 2022-08-15T16:15:53Z

Codecov Report

Merging #6639 (454aa4c) into main (235e4fe) will decrease coverage by 0.02%.
Report is 3 commits behind head on main.
The diff coverage is 0.00%.

@@             Coverage Diff              @@
##               main    #6639      +/-   ##
============================================
- Coverage     29.15%   29.13%   -0.02%     
+ Complexity    30909    30884      -25     
============================================
  Files          5165     5165              
  Lines        364111   364202      +91     
  Branches      53306    53306              
============================================
- Hits         106153   106119      -34     
- Misses       243397   243519     +122     
- Partials      14561    14564       +3

Flag	Coverage Δ
simulator-marvin-tests	`25.10% <0.00%> (-0.03%)`	⬇️
uitests	`4.53% <ø> (ø)`
unit-tests	`14.74% <ø> (ø)`

Flags with carried forward coverage won't be shown. Click here to find out more.

Files	Coverage Δ
ui/src/views/offering/AddComputeOffering.vue	`0.00% <ø> (ø)`
api/src/main/java/com/cloud/gpu/GPU.java	`2.77% <0.00%> (-14.87%)`	⬇️

... and 30 files with indirect coverage changes

📣 We’re building smart automated test selection to slash your CI/CD build times. Learn more

shwstppr

code LGTM.

In UI, however, with new options for GPU interface looks not so good. Radio button could be converted to select item,

-          <a-radio-group
+          <a-select
             v-model:value="form.pcidevice"
-            buttonStyle="solid"
-            @change="selected => { handleGpuChange(selected.target.value) }">
-            <a-radio-button v-for="(opt, optIndex) in gpuTypes" :key="optIndex" :value="opt.value">
+            showSearch
+            optionFilterProp="label"
+            :filterOption="(input, option) => {
+              return option.children[0].children.toLowerCase().indexOf(input.toLowerCase()) >= 0
+            }"
+            :placeholder="$t('label.gpu')"
+            @change="handleGpuChange">
+            <a-select-option v-for="(opt, optIndex) in gpuTypes" :key="optIndex" :value="opt.value">
               {{ opt.title }}
-            </a-radio-button>
-          </a-radio-group>
+            </a-select-option>
+          </a-select>

With select,

pdion891 · 2022-08-16T11:13:02Z

@DaanHoogland , I didn't test this yet, this is just a backport from previous version / UI.
I'll comment once tested and remove the PR from draft mode.

shwstppr · 2022-08-18T08:36:53Z

Accidentally marked this ready for review yesterday, moved it back to draft

yadvr · 2023-04-19T07:33:14Z

ping @pdion891 @shwstppr is this still relevant?

pdion891 · 2023-05-25T18:40:11Z

@rohityadavcloud yes, but at this point it would make sense to update it with recent GPU model.
I will rebase this PR and add some more GPU to it.

shwstppr · 2023-10-03T09:27:28Z

@pdion891 any update on this?

shwstppr · 2023-10-30T16:18:02Z

@blueorangutan ui

blueorangutan · 2023-10-30T16:20:04Z

@shwstppr a Jenkins job has been kicked to build UI QA env. I'll keep you posted as I make progress.

blueorangutan · 2023-10-30T16:33:45Z

UI build: ✔️
Live QA URL: https://qa.cloudstack.cloud/simulator/pr/6639 (QA-JID-213)

DaanHoogland · 2023-10-31T10:47:42Z

@pdion891 I updated the branch. Can you comment on the state of this PR?

DaanHoogland · 2023-10-31T10:47:51Z

@blueorangutan package

blueorangutan · 2023-10-31T10:48:03Z

@DaanHoogland a [SL] Jenkins job has been kicked to build packages. It will be bundled with KVM, XenServer and VMware SystemVM templates. I'll keep you posted as I make progress.

blueorangutan · 2023-10-31T11:51:50Z

Packaging result [SF]: ✔️ el7 ✔️ el8 ✔️ el9 ✔️ debian ✔️ suse15. SL-JID 7574

DaanHoogland · 2023-10-31T13:58:21Z

@blueorangutan test

blueorangutan · 2023-10-31T14:00:04Z

@DaanHoogland a [SL] Trillian-Jenkins test job (centos7 mgmt + kvm-centos7) has been kicked to run smoke tests

pdion891 · 2023-10-31T14:10:46Z

@shwstppr I've added your snippet suggestion, and thank you, it work great!

Here are some highlight about this PR: it remove vGPU support from legacy grid card not supported by recent NVIDIA driver.
because of the passthrough generic, any GPU passthrough is working but vGPU need to be defined.

This PR include known vGPU offering for A2, A10, A40, A5500, V100, T4 GPU card.

Some tests where done on XenServer 8.2cu1 with nvidia RTX driver 16.1. Not all vGPU definition have been tested, only made sure vGPU definition exist on XenServers.

DaanHoogland

clgtm, just wondering if encoding vGPUs in the UI is a good thing?

pdion891 · 2023-11-01T11:48:12Z

clgtm, just wondering if encoding vGPUs in the UI is a good thing?

it's not, it prevent us to easily add new GPU support. Currently, any GPU offering in passthrough mode can be created thru the API, but cannot thru the UI, they must be define like in this PR. ideally we would be able to define them a bit like GuestOS type works.

shwstppr · 2023-11-01T11:53:50Z

@blueorangutan ui

blueorangutan · 2023-11-01T11:54:04Z

@shwstppr a Jenkins job has been kicked to build UI QA env. I'll keep you posted as I make progress.

DaanHoogland · 2023-11-01T11:58:20Z

@shwstppr this has backend changes as well. Do you want to just test the UI?

blueorangutan · 2023-11-01T12:08:10Z

UI build: ✔️
Live QA URL: https://qa.cloudstack.cloud/simulator/pr/6639 (QA-JID-216)

shwstppr · 2023-11-01T12:08:35Z

@DaanHoogland missed those

@blueorangutan package

blueorangutan · 2023-11-01T12:10:03Z

@shwstppr a [SL] Jenkins job has been kicked to build packages. It will be bundled with KVM, XenServer and VMware SystemVM templates. I'll keep you posted as I make progress.

blueorangutan · 2023-11-01T13:08:31Z

Packaging result [SF]: ✔️ el7 ✔️ el8 ✔️ el9 ✔️ debian ✔️ suse15. SL-JID 7599

DaanHoogland · 2023-11-01T14:13:59Z

@blueorangutan test keepEnv

blueorangutan · 2023-11-01T14:16:04Z

@DaanHoogland a [SL] Trillian-Jenkins test job (centos7 mgmt + kvm-centos7) has been kicked to run smoke tests

blueorangutan · 2023-11-02T04:05:35Z

[SF] Trillian test result (tid-8205)
Environment: kvm-centos7 (x2), Advanced Networking with Mgmt server 7
Total time taken: 48274 seconds
Marvin logs: https://github.com/blueorangutan/acs-prs/releases/download/trillian/pr6639-t8205-kvm-centos7.zip
Smoke tests completed. 115 look OK, 1 have errors, 0 did not run
Only failed and skipped tests results shown below:

Test	Result	Time (s)	Test File
test_02_upgrade_kubernetes_cluster	`Failure`	579.13	test_kubernetes_clusters.py

DaanHoogland

looks good

and then

I think the way that vgpu is decided on needs an overhaul but would pull that out of scope of this PR

shwstppr

LGTM

boring-cyborg Bot added component:api component:UI labels Aug 15, 2022

pdion891 requested a review from shwstppr August 15, 2022 14:09

shwstppr reviewed Aug 16, 2022

View reviewed changes

shwstppr marked this pull request as ready for review August 17, 2022 04:15

shwstppr marked this pull request as draft August 18, 2022 08:35

DaanHoogland added this to the 4.19.0.0 milestone Jun 22, 2023

pdion891 and others added 5 commits October 31, 2023 11:47

Update GPU support for V100,T4,A5500.

2d9d65e

@shwstppr suggested UI change to use dropdown select box.

c8ac397

add vGPU and card support, fix ordering

09e4ca5

add vGPU types

375c982

remove outdated vGPU

454aa4c

DaanHoogland force-pushed the main-gpu-update branch from ef5d3f6 to 454aa4c Compare October 31, 2023 10:47

pdion891 marked this pull request as ready for review October 31, 2023 14:24

DaanHoogland reviewed Nov 1, 2023

View reviewed changes

DaanHoogland approved these changes Nov 2, 2023

View reviewed changes

shwstppr approved these changes Nov 2, 2023

View reviewed changes

DaanHoogland merged commit a3565fc into apache:main Nov 2, 2023

pdion891 deleted the main-gpu-update branch November 2, 2023 15:40

Conversation

pdion891 commented Aug 15, 2022 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Description

Types of changes

Feature/Enhancement Scale or Bug Severity

Feature/Enhancement Scale

How Has This Been Tested?

Uh oh!

acs-robot commented Aug 15, 2022

Uh oh!

blueorangutan commented Aug 15, 2022

Uh oh!

blueorangutan commented Aug 15, 2022

Uh oh!

DaanHoogland commented Aug 15, 2022

Uh oh!

blueorangutan commented Aug 15, 2022

Uh oh!

sonarqubecloud Bot commented Aug 15, 2022

Uh oh!

blueorangutan commented Aug 15, 2022

Uh oh!

DaanHoogland commented Aug 15, 2022

Uh oh!

codecov Bot commented Aug 15, 2022 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Codecov Report

Uh oh!

shwstppr left a comment

Choose a reason for hiding this comment

Uh oh!

pdion891 commented Aug 16, 2022

Uh oh!

shwstppr commented Aug 18, 2022

Uh oh!

yadvr commented Apr 19, 2023

Uh oh!

pdion891 commented May 25, 2023

Uh oh!

shwstppr commented Oct 3, 2023

Uh oh!

shwstppr commented Oct 30, 2023

Uh oh!

blueorangutan commented Oct 30, 2023

Uh oh!

blueorangutan commented Oct 30, 2023

Uh oh!

DaanHoogland commented Oct 31, 2023

Uh oh!

DaanHoogland commented Oct 31, 2023

Uh oh!

blueorangutan commented Oct 31, 2023

Uh oh!

blueorangutan commented Oct 31, 2023

Uh oh!

DaanHoogland commented Oct 31, 2023

Uh oh!

blueorangutan commented Oct 31, 2023

Uh oh!

pdion891 commented Oct 31, 2023 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

DaanHoogland left a comment

Choose a reason for hiding this comment

Uh oh!

pdion891 commented Nov 1, 2023

Uh oh!

shwstppr commented Nov 1, 2023

Uh oh!

blueorangutan commented Nov 1, 2023

Uh oh!

DaanHoogland commented Nov 1, 2023

Uh oh!

blueorangutan commented Nov 1, 2023

Uh oh!

shwstppr commented Nov 1, 2023

Uh oh!

blueorangutan commented Nov 1, 2023

Uh oh!

pdion891 commented Aug 15, 2022 •

edited

Loading

codecov Bot commented Aug 15, 2022 •

edited

Loading

pdion891 commented Oct 31, 2023 •

edited

Loading