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.

Sign up for GitHub

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

Jump to bottom

Improve end to end watch node event latency -- currently prolonged ( > 1 sec ) is high #83

Closed
q131172019 opened this issue Jul 13, 2022 · 3 comments
Closed

Improve end to end watch node event latency -- currently prolonged ( > 1 sec ) is high #83

q131172019 opened this issue Jul 13, 2022 · 3 comments
Milestone
2022-0930

Comments

@q131172019
Copy link
Collaborator

In https://github.com/yb01/arktos/wiki/730-test, for Test case - 500K nodes / 2 regions / 20 schedulers / 25K nodes per scheduler, when one resource region manager simulator is in AWS region us-west-1, another resource region manager simulator, service-api, all schedulers are in AWS region us-west-2,
The percentages of "Number of nodes watch prolonged than 1s" on all schedulers reach 20-40% high dramatically.

It is wired.
@q131172019 q131172019 changed the title One region simulator in different AWS region from that of service-api, schedulers and another region simulator causes high percentage of "Number of nodes watch prolonged than 1s" One region simulator in different AWS region from that of service-api, schedulers and another region simulator causes high percentage of "Number of nodes watch prolonged than 1s" on all schedulers Jul 13, 2022
@yb01 yb01 added this to the 2022-0730 milestone Jul 14, 2022
@yb01
Copy link
Collaborator

yb01 commented Jul 14, 2022

must fix for 730.

@yb01 yb01 modified the milestones: 2022-0730, 2022-0930 Aug 2, 2022
@yb01
Copy link
Collaborator

yb01 commented Aug 8, 2022 •
edited
Loading

keep it for 930 now. likely move to 1230 as perf goal. --- RP down data change

@yb01 yb01 changed the title One region simulator in different AWS region from that of service-api, schedulers and another region simulator causes high percentage of "Number of nodes watch prolonged than 1s" on all schedulers Improve end to end watch node event latency -- currently prolonged ( > 1 sec ) is high Aug 11, 2022
@yb01
Copy link
Collaborator

yb01 commented Oct 14, 2022

no repro anymore

@yb01 yb01 closed this as completed Oct 14, 2022
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
2022-0930
Development

No branches or pull requests
2 participants
@yb01 @q131172019