[ci] Deploy profiler in GitHub using large runner

single-cell-data / TileDB-SOMA

Python and R SOMA APIs using TileDB’s cloud-native format. Ideal for single-cell data at any scale.

https://tiledbsoma.readthedocs.io

MIT License

92 stars 25 forks source link

[ci] Deploy profiler in GitHub using large runner #1607

Open beroy opened 1 year ago

beroy commented 1 year ago

[parent #1600] In order for the profiler to be really effective, it needs to be embed in our Github flow to trigger the right performance tests for each major release or patch. Addressed in https://github.com/single-cell-data/TileDB-SOMA/pull/1840

beroy commented 1 year ago

Few candidates for next steps:

Platform:

Github actions —> not the best choice
GH runner with custom runner
Do we even do it in CZI under tileDB SOMA or TileDB
If we can’t we can provide documentation on how exactly to do documentation
EC2 instance
SLOs —> where we were last time

Benchmarks:

Experiment Axis Query test ON Reference data not on CellCensus
Benchmark generate data and write it and then another test read it