Open quasiben opened 2 weeks ago
_See test report for an extended history of previous test failures. This is useful for diagnosing flaky tests._
29 files ± 0 29 suites ±0 11h 4m 19s :stopwatch: - 27m 20s 4 064 tests + 4 3 958 :white_check_mark: +3 101 :zzz: + 4 5 :x: - 1 55 941 runs +15 53 756 :white_check_mark: +4 2 178 :zzz: +15 7 :x: - 2
For more details on these failures, see this check.
Results for commit 4e963b8d. ± Comparison against base commit af237f0d.
I can, but with PyData London tomorrow I'm not going to get to this immediately.
I've tested this on machine with several A100s. @jacobtomlinson is it possible for you to test this PR in a MIG setup ?
@jacobtomlinson: Gentle ping :)
Apologies I've been struggling to get my hands on the right hardware to test this out. Let me ping NVIDIA folks again.
Apologies I've been struggling to get my hands on the right hardware to test this out. Let me ping NVIDIA folks again.
No worries, I just wanted to make sure this didn't get buried in the post-conference catch-up.
Replaces https://github.com/dask/distributed/pull/8419
cc @fjetter @hendrikmakait @rjzamora
I've tested this on machine with several A100s. @jacobtomlinson is it possible for you to test this PR in a MIG setup ?