[Test] Cap OSU collective message size at 8K on burstable-network instances - #7619
Closed
hanwen-cluster wants to merge 1 commit into
Closed
hanwen-cluster wants to merge 1 commit into
hanwen-cluster wants to merge 1 commit into
Conversation
hanwen-cluster
force-pushed
the
developsep10
branch
2 times, most recently
from
September 10, 2026 20:22
444f0cf to
35674db
Compare
gmarciani
reviewed
Sep 10, 2026
| assert_no_errors_in_logs(remote_command_executor, scheduler, skip_ice=True) | ||
|
|
||
|
|
||
| def _has_burstable_network(instance_info): |
Contributor
There was a problem hiding this comment.
[minor] why two loops on cards?
Can't we just do a single loop with something like?
for card in network_info.get("NetworkCards"):
if "BaselineBandwidthInGbps" in card and "PeakBandwidthInGbps" in card:
return card["BaselineBandwidthInGbps"] < card["PeakBandwidthInGbps"]
| CONFIRMATION_REPETITIONS = 2 | ||
|
|
||
| # Largest collective message size run on instance types with a burstable network. | ||
| BURSTABLE_NETWORK_MAX_MESSAGE_SIZE = 8192 |
Contributor
There was a problem hiding this comment.
How did we come with this value?
Doesn't it depend on the specific bandwidth cap of the instance type?
…IXES We were not correctly excluding instance types like g2, and generated failures in test_essential_features in some regions
hanwen-cluster
force-pushed
the
developsep10
branch
from
September 25, 2026 13:25
35674db to
b32628d
Compare
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Description of changes
On 500 x c5.xlarge the collectives above 8K are bandwidth bound and run above the instance's 1.25 Gbps network baseline, so the result tracks EC2 burst credits rather than the software under test. In the perf history, those tests show most noises, and don't give clear signals of performance degradation
Tests
Checklist
developadd the branch name as prefix in the PR title (e.g.[release-3.6]).Please review the guidelines for contributing and Pull Request Instructions.
By submitting this pull request, I confirm that my contribution is made under the terms of the Apache 2.0 license.