Revamp End-to-End testing

lablup / backend.ai

Backend.AI is a streamlined, container-based computing cluster platform that hosts popular computing/ML frameworks and diverse programming languages, with pluggable heterogeneous accelerator support including CUDA GPU, ROCm GPU, TPU, IPU and other NPUs.

GNU Lesser General Public License v3.0

505 stars 150 forks source link

Main idea

Since Backend.AI project is getting complicated even more these days, it is almost impossible for us to test core functionalities only with our hands. Therefore I think this is the right time to revamp our old fashioned integration test package and introduce a daily E2E testing CI pipeline.

[x] Update (or rewrite) ai.backend.test.cli_integration package so that it can be ran and report errors
[ ] Add new test cases for other uncovered CLI features
[ ] Create a Github action pipeline to execute the integration on a daily basis

Alternative ideas

No response

Anything else?

No response

lablup / backend.ai