Client for FAH resources, reference implementation for compute service

OpenFreeEnergy / alchemiscale-fah

protocols and compute service for using alchemiscale with Folding@Home

MIT License

2 stars 0 forks source link

Client for FAH resources, reference implementation for compute service #7

Closed dotsdl closed 3 months ago

dotsdl commented 1 year ago

Closes #1, #3.

codecov-commenter commented 8 months ago

Welcome to Codecov :tada:

Once you merge this PR into your default branch, you're all set! Codecov will compare coverage reports and display results in all future pull requests.

Thanks for integrating Codecov - We've got you covered :open_umbrella:

dotsdl commented 6 months ago

@ianmkenney, since this is such a monster PR, can I get an initial review from you? This will help us identify gaps in testing, clarity, etc. as I finish up the test suite for the compute service and the CLI. It will also get you oriented enough with what's here to assist me in troubleshooting issues as I finish out those tests.

dotsdl commented 5 months ago

Looks like CI is now pulling in openfe and gufe 1.0, which alchemiscale and feflow are not yet compatible with. I'll proceed with pinning our test and deployment env to what we're currently using in alchemiscale so we aren't blocked in doing deployment test cycles.

dotsdl commented 5 months ago

CI passes! Now working on:

[x] CLI entrypoint for service startup
[x] YAML config template for service
[x] tests for CLI entrypoint
[x] tests for cert renewal functionality

dotsdl commented 4 months ago

Thank you @ianmkenney and @hmacdope! These reviews are extremely helpful! I'm making my way through your recommendations!

dotsdl commented 4 months ago

Note for self: need to make sure we add a file-based indicator to completed RUNs/GENs so that a separate archive/cleanup service can consume these.

dotsdl commented 4 months ago

Also, during my deployment testing I realized we've made an oversight in the design of how we interface ProtocolDAG execution with the FAH PRC(G) system. Instead of a Task mapping to a CLONE, we should instead have a ProtocolUnit map to a CLONE, since a given Task's ProtocolDAG may in principle feature any number of ProtocolUnits that are FahSimulationUnits. This means that for a single Task, it's possible multiple FAH CLONEs will be performed.

Fixing this isn't too difficult given how we've laid things out, but making this adjustment will require changes in a few places.

dotsdl commented 3 months ago

Addressed this today; we now use Task-ProtocolUnit pairs to map into FAH CLONES instead of just Tasks.

hmacdope commented 3 months ago

Congrats @dotsdl