Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Implement GPU Support #728

Merged
merged 11 commits into from
Dec 12, 2024

Conversation

nesitor
Copy link
Member

@nesitor nesitor commented Dec 3, 2024

Related Jira tickets : ALEPH-314

Self proofreading checklist

  • The new code clear, easy to read and well commented.
  • New code does not duplicate the functions of builtin or popular libraries.
  • An LLM was used to review the new code and look for simplifications.
  • New classes and functions contain docstrings explaining what they provide.
  • All new code is covered by relevant tests.
  • Documentation has been updated regarding these changes.
  • Dependencies update in the project.toml have been mirrored in the Debian package build script packaging/Makefile

How to test

Create a QEmu VM with assigned GPUs and check that you can use it inside the Qemu VM

Solution: Implement GPU assignation feature that will be pass-though to QEmu VMs with native performance.
@nesitor nesitor self-assigned this Dec 3, 2024
@nesitor nesitor changed the base branch from main to proto-add-gpu-listing December 3, 2024 21:39
@nesitor nesitor force-pushed the andres-feature-implement_gpu_support branch from a9f85eb to 24516a2 Compare December 5, 2024 12:29
@nesitor nesitor force-pushed the andres-feature-implement_gpu_support branch from 891ed73 to 60b2491 Compare December 5, 2024 17:47
Copy link

codecov bot commented Dec 5, 2024

Codecov Report

Attention: Patch coverage is 40.74074% with 64 lines in your changes missing coverage. Please review.

Project coverage is 62.46%. Comparing base (fb379ff) to head (11141af).
Report is 1 commits behind head on proto-add-gpu-listing.

Files with missing lines Patch % Lines
...5c6ae643c69b_add_gpu_column_to_executions_table.py 0.00% 20 Missing ⚠️
src/aleph/vm/models.py 45.83% 13 Missing ⚠️
src/aleph/vm/pool.py 20.00% 12 Missing ⚠️
src/aleph/vm/orchestrator/tasks.py 14.28% 6 Missing ⚠️
src/aleph/vm/orchestrator/views/__init__.py 0.00% 6 Missing ⚠️
src/aleph/vm/hypervisors/qemu/qemuvm.py 86.66% 1 Missing and 1 partial ⚠️
...c/aleph/vm/hypervisors/qemu_confidential/qemuvm.py 0.00% 2 Missing ⚠️
src/aleph/vm/orchestrator/cli.py 0.00% 2 Missing ⚠️
src/aleph/vm/orchestrator/chain.py 66.66% 1 Missing ⚠️
Additional details and impacted files
@@                    Coverage Diff                    @@
##           proto-add-gpu-listing     #728      +/-   ##
=========================================================
- Coverage                  62.79%   62.46%   -0.33%     
=========================================================
  Files                         72       73       +1     
  Lines                       6343     6424      +81     
  Branches                     514      528      +14     
=========================================================
+ Hits                        3983     4013      +30     
- Misses                      2199     2249      +50     
- Partials                     161      162       +1     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

@nesitor nesitor force-pushed the andres-feature-implement_gpu_support branch from 0c58cad to ba1fc9d Compare December 6, 2024 10:21
Copy link
Collaborator

@olethanh olethanh left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I already add a few comments and concerns. I haven't been able to test it fully yet.

  • aleph-vm refuse to launch properly as it is missing migration for the newly added database column. This is going to be a problem in prod.
  • The test server I use at the moment don't have a GPU, it's going to be added on Thursday.

src/aleph/vm/hypervisors/qemu/qemuvm.py Outdated Show resolved Hide resolved
src/aleph/vm/orchestrator/tasks.py Outdated Show resolved Hide resolved
src/aleph/vm/orchestrator/metrics.py Show resolved Hide resolved
src/aleph/vm/orchestrator/payment.py Outdated Show resolved Hide resolved
@nesitor nesitor requested a review from olethanh December 9, 2024 18:35
@nesitor nesitor force-pushed the andres-feature-implement_gpu_support branch from c9b717e to 70b7390 Compare December 9, 2024 18:40
@olethanh olethanh force-pushed the andres-feature-implement_gpu_support branch from 6a12504 to d919116 Compare December 9, 2024 20:06
@olethanh
Copy link
Collaborator

I have tested with the CRN Andres put a my disposal and it seems to be working

@nesitor nesitor force-pushed the andres-feature-implement_gpu_support branch from 5f2579a to 11141af Compare December 11, 2024 20:11
@nesitor nesitor merged commit bb5d74d into proto-add-gpu-listing Dec 12, 2024
18 of 20 checks passed
@nesitor nesitor deleted the andres-feature-implement_gpu_support branch December 12, 2024 14:25
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants