Bug 1310180

Summary:	[RFE] IO threads should be considered in scheduling
Product:	[oVirt] ovirt-engine	Reporter:	Milan Zamazal <mzamazal>
Component:	General	Assignee:	bugs <bugs>
Status:	CLOSED DEFERRED	QA Contact:
Severity:	low	Docs Contact:
Priority:	unspecified
Version:	4.0.0	CC:	bugs, dfediuck, mgoldboi, michal.skrivanek, mzamazal
Target Milestone:	---	Keywords:	FutureFeature, Improvement
Target Release:	---	Flags:	mgoldboi: needinfo? (mgoldboi) dfediuck: ovirt-future? rule-engine: planning_ack? rule-engine: devel_ack? rule-engine: testing_ack?
Hardware:	Unspecified
OS:	Unspecified
Whiteboard:
Fixed In Version:		Doc Type:	Enhancement
Doc Text:		Story Points:	---
Clone Of:		Environment:
Last Closed:	2020-04-01 14:48:03 UTC	Type:	Bug
Regression:	---	Mount Type:	---
Documentation:	---	CRM:
Verified Versions:		Category:	---
oVirt Team:	SLA	RHEL 7.3 requirements from Atomic Host:
Cloudforms Team:	---	Target Upstream Version:

Description Milan Zamazal 2016-02-19 16:44:27 UTC

Description of problem:

Support for IO threads (AKA virtio-blk-dataplane) was added to oVirt some time ago, see https://bugzilla.redhat.com/1214311. This functionality utilizes dedicated threads and thus the number of available CPU cores should be taken into account. AFAIK this is not considered in oVirt scheduling.

Actual results:

Possibly suboptimal resource usage due to higher requirements on the number of CPU cores than expected.

Possible solution:

Inclusion of the number of requested IO threads and the corresponding system resource requirements into scheduling calculations.

Comment 1 Doron Fediuck 2016-02-21 08:05:07 UTC

Can you be more specific about the request;
ie- which calculation should consider the IO threads and in what way?

Comment 2 Milan Zamazal 2016-02-23 10:28:59 UTC

I don't know which CPU related scheduling calculations should be affected, AFAIK IO threads are not counted anywhere.  Initial clue to how to count them is in libvirt documentation: "There should be only 1 or 2 IOThreads per host CPU."

Comment 3 Moran Goldboim 2016-11-30 10:22:12 UTC

please check out with performance team.

Comment 5 Michal Skrivanek 2020-03-19 15:41:36 UTC

We didn't get to this bug for more than 2 years, and it's not being considered for the upcoming 4.4. It's unlikely that it will ever be addressed so I'm suggesting to close it.
If you feel this needs to be addressed and want to work on it please remove cond nack and target accordingly.

Comment 6 Michal Skrivanek 2020-04-01 14:48:03 UTC

ok, closing. Please reopen if still relevant/you want to work on it.

Comment 7 Michal Skrivanek 2020-04-01 14:51:22 UTC

ok, closing. Please reopen if still relevant/you want to work on it.