It would be great to find a way to make these benchmarks get run somewhere in the cloud every time we merge a PR on this repo. The basic idea would be:
Travis CI fires up as usual
Travis tells GCP / AWS to fire up some virtual machines (do we need a docker image?)
Those VMs clone the repo and run asv
asv automatically creates the appropriate conda environments (that is already part of how asv works)
a script posts the results in some persistent location
I don't know enough cloud juju to figure out how to do all this. But it would be cool.
It would be great to find a way to make these benchmarks get run somewhere in the cloud every time we merge a PR on this repo. The basic idea would be:
I don't know enough cloud juju to figure out how to do all this. But it would be cool.