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

mpm_event module to improve apache scalability #3516

Merged
merged 1 commit into from
Nov 17, 2023

Conversation

athithyaaselvam
Copy link
Collaborator

@athithyaaselvam athithyaaselvam commented Oct 30, 2023

What changes were proposed in this pull request?

This pull request addresses the throttling of concurrent connections (approximately at 200 simultaneous threads) by the Apache load balancer when directing traffic to the Hue server. The scalability of the Apache load balancer is enhanced by implementing the mpm_event module.

How was this patch tested?

  • The patch underwent stress testing using Jmeter.

  • Tests were conducted on both Redhat8 and Centos7 systems with varying computational capacities and node configurations.

  • Below is a sample result from a Redhat 8 machine:

    Default Apache Configurations:

  "Summary: + 1591 in 00:00:30 = 53.1/s Avg: 20938 Min: 0 Max: 172930 Err: 510 (32.06%) Active: 989 Started: 1000 Finished: 11
  Summary Total: 14118 in 00:09:17 = 25.4/s Avg: 20088 Min: 0 Max: 174866 Err: 2668 (18.90%)"
  • The test concluded in roughly 10 minutes, although it was anticipated to run for over 3 hours. This early completion was due to the Apache Load Balancer throttling connections, preventing responses from the Hue server.

    With mpm_event Configuration Changes and an Extended KeepAliveTimeout:

  "Summary: + 395 in 00:00:30 = 13.0/s Avg: 74657 Min: 101 Max: 249963 Err: 4 (1.01%) Active: 1000 Started: 1000 Finished: 0
  Summary Total: 81274 in 01:41:19 = 13.4/s Avg: 70975 Min: 0 Max: 690747 Err: 1863 (2.29%)"
  • The test had a duration of approximately 1 hour and 40 minutes (was manually killed). Remarkably, none of the 1000 concurrent threads were terminated, and all received responses from the Hue server.

  • The error rate with the default Apache load balancer configuration stood at 18%. However, after implementing the mpm_event module configurations, the error rate significantly dropped to around 2%.

Please review Hue Contributing Guide before opening a pull request.

@athithyaaselvam athithyaaselvam force-pushed the apache_lb_bottleneck branch 2 times, most recently from ae299f5 to 1a4eb49 Compare October 30, 2023 23:55
Copy link
Collaborator

@agl29 agl29 left a comment

Choose a reason for hiding this comment

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

LGTM!

Copy link
Collaborator

@wing2fly wing2fly left a comment

Choose a reason for hiding this comment

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

please remove the change-id in the git comment

@athithyaaselvam athithyaaselvam enabled auto-merge (squash) November 17, 2023 00:34
@athithyaaselvam athithyaaselvam merged commit 10a46d1 into master Nov 17, 2023
2 checks passed
@athithyaaselvam athithyaaselvam deleted the apache_lb_bottleneck branch November 17, 2023 22:04
athithyaaselvam added a commit that referenced this pull request Feb 5, 2024
Change-Id: I2ec0dbe6c07a0c510a5f9c43a1b389630464d625

Co-authored-by: Athithyaa Selvam <aselvam@cloudera.com>
(cherry picked from commit 10a46d1)
Alterrien pushed a commit to Alterrien/hue that referenced this pull request Dec 19, 2024
Change-Id: I2ec0dbe6c07a0c510a5f9c43a1b389630464d625

Co-authored-by: Athithyaa Selvam <aselvam@cloudera.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.

4 participants