Minor changes for snappi_tests for T2-ixia, change udp port calculations.#15906
Merged
yejianquan merged 4 commits intosonic-net:masterfrom Dec 11, 2024
Merged
Conversation
Contributor
|
@rraghav-cisco can you separate this PR into two? one for the fixture change, another one for the UDP port and flow rate change. |
zhixzhu
approved these changes
Dec 5, 2024
Contributor
|
@selldinesh @amitpawar12 for review |
rraghav-cisco
changed the title
alpeshspatel
approved these changes
Dec 6, 2024
Contributor
Author
Contributor
Contributor
|
@rraghav-cisco can you update the description with more detail on why these changes are needed? and test result as well? |
Contributor
Author
|
results: <style> </style>
|
mssonicbld
pushed a commit
to mssonicbld/sonic-mgmt
that referenced
this pull request
Dec 11, 2024
…ons. (sonic-net#15906) Description of PR Currently the traffic for RDMA tests use randomly selected udp ports. This sometimes ends up sending traffic in a non-load-balanced way, and causes the tests to fail. In this PR: For multiple scripts that use udp: We serialize the UDP port numbers instead of randomly picking the numbers. This allows for more uniform load-balancing spread, and allows tests to pass more. in multi_lossless helper: We reduce the background traffic rate, and raise the rate for lossless traffic, thus keeping the overall rate still same. This allows for accounting the speed of backplane. co-authorized by: jianquanye@microsoft.com
Collaborator
|
Cherry-pick PR to 202405: #15990 |
mssonicbld
pushed a commit
that referenced
this pull request
Dec 11, 2024
…ons. (#15906) Description of PR Currently the traffic for RDMA tests use randomly selected udp ports. This sometimes ends up sending traffic in a non-load-balanced way, and causes the tests to fail. In this PR: For multiple scripts that use udp: We serialize the UDP port numbers instead of randomly picking the numbers. This allows for more uniform load-balancing spread, and allows tests to pass more. in multi_lossless helper: We reduce the background traffic rate, and raise the rate for lossless traffic, thus keeping the overall rate still same. This allows for accounting the speed of backplane. co-authorized by: jianquanye@microsoft.com
mssonicbld
added
Included in 202405 branch
and removed
Created PR to 202405 branch
labels
sdszhang
reviewed
Dec 11, 2024
| TEST_FLOW_AGGR_RATE_PERCENT = 30 | ||
| TEST_FLOW_AGGR_RATE_PERCENT = 35 | ||
| BG_FLOW_NAME = 'Background Flow' | ||
| BG_FLOW_AGGR_RATE_PERCENT = 25 |
Contributor
There was a problem hiding this comment.
In the description:
"in multi_lossless helper: We reduce the background traffic rate, and raise the rate for lossless traffic, thus keeping the overall rate still same. This allows for accounting the speed of backplane."
Earlier lossless + lossy traffic is 30% + 25% = 55% from each ingress.
Now lossless + lossy traffic is 35% + 22.5% = 57.5% from each ingress.
The traffic rate is not same.
Contributor
Author
There was a problem hiding this comment.
Yes, @amitpawar12 . That was intentional. We had discussions, and since lossless will get paused anyways, it can be higher, if we want.
mssonicbld
pushed a commit
to mssonicbld/sonic-mgmt
that referenced
this pull request
Jan 2, 2025
…ons. (sonic-net#15906) Description of PR Currently the traffic for RDMA tests use randomly selected udp ports. This sometimes ends up sending traffic in a non-load-balanced way, and causes the tests to fail. In this PR: For multiple scripts that use udp: We serialize the UDP port numbers instead of randomly picking the numbers. This allows for more uniform load-balancing spread, and allows tests to pass more. in multi_lossless helper: We reduce the background traffic rate, and raise the rate for lossless traffic, thus keeping the overall rate still same. This allows for accounting the speed of backplane. co-authorized by: jianquanye@microsoft.com
Collaborator
|
Cherry-pick PR to 202411: #16294 |
mssonicbld
pushed a commit
that referenced
this pull request
Jan 3, 2025
…ons. (#15906) Description of PR Currently the traffic for RDMA tests use randomly selected udp ports. This sometimes ends up sending traffic in a non-load-balanced way, and causes the tests to fail. In this PR: For multiple scripts that use udp: We serialize the UDP port numbers instead of randomly picking the numbers. This allows for more uniform load-balancing spread, and allows tests to pass more. in multi_lossless helper: We reduce the background traffic rate, and raise the rate for lossless traffic, thus keeping the overall rate still same. This allows for accounting the speed of backplane. co-authorized by: jianquanye@microsoft.com
mssonicbld
added
Included in 202411 branch
and removed
Created PR to 202411 branch
labels
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
8 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 PR
Currently the traffic for RDMA tests use randomly selected udp ports. This sometimes ends up sending traffic in a non-load-balanced way, and causes the tests to fail.
In this PR:
Type of change
Back port request