Bug 1965928

Summary:	performance search rate: contention on global monitoring counters
Product:	Red Hat Enterprise Linux 8	Reporter:	thierry bordaz <tbordaz>
Component:	389-ds-base	Assignee:	thierry bordaz <tbordaz>
Status:	CLOSED ERRATA	QA Contact:	RHDS QE <ds-qe-bugs>
Severity:	unspecified	Docs Contact:
Priority:	unspecified
Version:	8.3	CC:	bsmejkal, gkimetto, ldap-maint, mreynolds, pcech, sgouvern
Target Milestone:	beta	Keywords:	Triaged
Target Release:	8.7
Hardware:	Unspecified
OS:	Unspecified
Whiteboard:	sync-to-jira
Fixed In Version:	389-ds-base-1.4.3.30-3.module+el8.7.0+15600+347cafc6	Doc Type:	If docs needed, set a value
Doc Text:	DS is a multi-threaded application. Several threads (workers and listeners) a updating common counters used for monitoring. Under high workload, several threads can access at the same time the same counters creating a kind of contention (cache lines moving from one CPU to another). The fix prevents that the threads access the same counters that eliminates contention.	Story Points:	---
Clone Of:		Environment:
Last Closed:	2022-11-08 09:38:12 UTC	Type:	Bug
Regression:	---	Mount Type:	---
Documentation:	---	CRM:
Verified Versions:		Category:	---
oVirt Team:	---	RHEL 7.3 requirements from Atomic Host:
Cloudforms Team:	---	Target Upstream Version:
Embargoed:

Description thierry bordaz 2021-05-31 07:30:18 UTC

Description of problem:
Many counters are related to cn=monitoring (snmp, ops, maxthread..). They are global to the server and listener/workers update them frequently. Those (atomic) update look expensive as skipping them improves throughput by >2%.
This ticket is to evaluate the gain of moving some of those global counters to a per workers/listener. Then on cn=monitor search, the per worker/listener counter would be consolidate into a global server counter.

Version-Release number of selected component (if applicable):


How reproducible:


Steps to Reproduce:
1.
2.
3.

Actual results:


Expected results:


Additional info:

Comment 2 thierry bordaz 2021-12-09 17:43:44 UTC

Fix is pushed upstream => POST

Comment 11 thierry bordaz 2022-06-10 12:01:34 UTC

Fixed in 389-ds-base-1.4.3.30-3.module+el8.7.0+15600+347cafc6 => MODIFIED

Comment 18 errata-xmlrpc 2022-11-08 09:38:12 UTC

Since the problem described in this bug report should be
resolved in a recent advisory, it has been closed with a
resolution of ERRATA.

For information on the advisory (389-ds:1.4 bug fix and enhancement update), and where to find the updated
files, follow the link below.

If the solution does not work for you, open a new bug report.

https://access.redhat.com/errata/RHBA-2022:7552