Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

http2: shard http2 codec_impl_test #5341

Merged
merged 2 commits into from
Dec 18, 2018

Conversation

akonradi
Copy link
Contributor

This test takes forever to run but the test cases are trivially parallelizable. With multiple cores, this reduces the running time significantly.

Risk Level: low
Testing: ran affected test
Docs Changes: n/a
Release Notes: n/a

This test takes forever to run but the test cases are trivially
parallelizable. With multiple cores, this reduces the running time
significantly.

Signed-off-by: Alex Konradi <akonradi@google.com>
Copy link
Contributor

@alyssawilk alyssawilk left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

+1 for sharding. Well honestly I'd prefer fewer combinations and permutations but I recall that getting push going from 1 to 50 seems a bit high - we have other fairly long running tests. Maybe just start with a few shards for first? What are you seeing for timing?

@alyssawilk alyssawilk self-assigned this Dec 18, 2018
@akonradi
Copy link
Contributor Author

Bazel uses 11 cores on my machine; under TSAN this takes 25 seconds sharded, 96 seconds unsharded (when the target is already built). With the regular build it takes 8 seconds sharded and 18 seconds unsharded. The motivation here is mostly to split up the target during execution since it uses way too much memory for our internal TSAN builds.

Too many shards causes slowdowns for single-core machines. Decrease the
shard count to keep this from being an issue for CI.

Signed-off-by: Alex Konradi <akonradi@google.com>
@mattklein123 mattklein123 merged commit 20b1930 into envoyproxy:master Dec 18, 2018
fredlas pushed a commit to fredlas/envoy that referenced this pull request Mar 5, 2019
This test takes forever to run but the test cases are trivially
parallelizable. With multiple cores, this reduces the running time
significantly.

Signed-off-by: Alex Konradi <akonradi@google.com>
Signed-off-by: Fred Douglas <fredlas@google.com>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

3 participants