A bit of polish on our nightly report

[jonkeane]

A few cleanups:

• Cleaned up our crossbow nightly report to be tabbed
  

• fixed machine names for our performance reports

• updating renv locks to actually install binaries

• cleaner names for GHA jobs

@assignUser @raulcd Do either of you know how the performance report gets triggered in our release process? I saw runs like https://github.com/ursacomputing/crossbow/actions/runs/14333506941 but I don't quite know what triggers that?

cc @boshek

[assignUser]

Oh this is great, thank you!

We trigger the workflow that generates the report manually from the actions tab with refs for base and contender that have existing conbench runs. I manually triggered a benchmark on the rc branch for it but ran into the issue with the machine names.

[jonkeane]

We trigger the workflow that generates the report manually from the actions tab with refs for base and contender that have existing conbench runs. I manually triggered a benchmark on the rc branch for it but ran into the issue with the machine names.

Ha! I just assumed this was all automatic. Is there a (more) automatic way of getting the baseline and contender shas? I actually found that was one of the most annoying parts of getting this setup to run again locally. Also, did I miss a reference to it in the release checklist/guide? I went looking for it but couldn't find anything. It's not a huge deal but would be good if that was discoverable :)

[assignUser]

Agreed on both points, it is quite annoying and should be made easier but at the minimum be added to the release guide! I'll open an Issue.

[assignUser]

apache/arrow#46323

[jonkeane]

Thanks 🙏