Preliminary Vast AI support

[kristopolous]

This is preliminary support for Vast. It currently works on an unreleased version of the SDK which we will soon get up to PyPy

The document https://docs.google.com/document/d/1oWox3qb3Kz3wXXSGg9ZJWwijoa99a3PIQUHBR8UgEGs/edit?pli=1&tab=t.0 was followed and all the testing passed

Tested (run the relevant ones):

• Code formatting: bash format.sh
• Any manual or new tests for this PR (please specify below)
• All smoke tests: pytest tests/test_smoke.py
• Relevant individual smoke tests: pytest tests/test_smoke.py::test_fill_in_the_name
• Backward compatibility tests: conda deactivate; bash -i tests/backward_compatibility_tests.sh

I'm pretty sure there will need to be edits, I'm fine with that. This is attempt 1. The outstanding work:

We need to

• tidy up our dockerhub and will get a better image to launch.
• release the updates to the sdk and come up with a pip name for it.
• get our catalog to update in the git hook flow as described (my goal is every 6 hours)

[cblmemo]

Thanks for contributing to this @kristopolous ! This is really exciting. Left some discussions. One main confusion I have is that, does vast ai like runpod, a cloud providing pods to users as their "VM"s? Asking because I'm seeing a lot of docker related code, and just want to confirm :)

[kristopolous]

Thanks for contributing to this @kristopolous ! This is really exciting. Left some discussions. One main confusion I have is that, does vast ai like runpod, a cloud providing pods to users as their "VM"s? Asking because I'm seeing a lot of docker related code, and just want to confirm :)

historically, runpod was a clone of vast. We currently offer docker-style containers and will be providing vms soonish (probably before end of year)

[kristopolous]

these test passing is blocked by https://github.com/skypilot-org/skypilot-catalog/pull/100/commits

[kristopolous]

We only offer GPUs instances ... you are free to use the CPU if you'd like but we're a GPU shop

Got it. Feel free to ignore this one.

so most of these are ok now ... i've done a number of fixes to the tests in general ... I can send you my logs but that's just "works on my machine" stuff ... so let my communicate how I'm running these:

10:25 /home/chris/code/skypilot$ source env/bin/activate                         
(env) 10:25 /home/chris/code/skypilot$ pip3 install -e .                                                                                                                                                             
(env) 10:25 /home/chris/code/skypilot$ pytest -v -n 1 tests/test_smoke.py --vast 

if we can agree this is sensible, all the tests you asked for should be passing

[cblmemo]

We only offer GPUs instances ... you are free to use the CPU if you'd like but we're a GPU shop

Got it. Feel free to ignore this one.

so most of these are ok now ... i've done a number of fixes to the tests in general ... I can send you my logs but that's just "works on my machine" stuff ... so let my communicate how I'm running these:

10:25 /home/chris/code/skypilot$ source env/bin/activate                         
(env) 10:25 /home/chris/code/skypilot$ pip3 install -e .                                                                                                                                                             
(env) 10:25 /home/chris/code/skypilot$ pytest -v -n 1 tests/test_smoke.py --vast 

if we can agree this is sensible, all the tests you asked for should be passing

Seems like there are still some CI test failing (tests/test_optimizer_random_dag.py. Could you help fixing them?

[cblmemo]

Thanks for fixing this @kristopolous ! It mostly looks good to me. Left final nits ;)

[cblmemo]

/smoke-test aws

[cblmemo]

Seems like the smoke test on AWS passed as well. After resolving those final nits it should be ready to go!

[kristopolous]

test_optimizer_random_dag

pytest -v -n 1 tests/test_optimizer_random_dag.py --vast

This test passes. Is there anything else we need?

[cblmemo]

Thanks for fixing @kristopolous ! Left one final nits. Could you also help resolve the merge conflicts?

Seems like the CI is failing because the catalog is missing. I just merged the catalog PR and I'm running the tests one last time. If it passed then this should be ready to go!

[cblmemo]

/smoke-test aws

[cblmemo]

Seems like the tests/test_optimizer_dryruns.py is failing. Could you help to fix that?

https://github.com/skypilot-org/skypilot/actions/runs/12971661591/job/36178313189?pr=4365