Liron Ar has posted comments on this change.

Change subject: core: long query response time while many simultaneously 
queries are running
......................................................................


Patch Set 10:

(2 comments)

http://gerrit.ovirt.org/#/c/27586/10/backend/manager/modules/vdsbroker/src/main/java/org/ovirt/engine/core/vdsbroker/VdsUpdateRunTimeInfo.java
File 
backend/manager/modules/vdsbroker/src/main/java/org/ovirt/engine/core/vdsbroker/VdsUpdateRunTimeInfo.java:

Line 1977:             vmToUpdate.updateRunTimeStatisticsData(vmStatistics, 
vmToUpdate);
Line 1978:             addVmStatisticsToList(vmToUpdate.getStatisticsData());
Line 1979:             updateInterfaceStatistics(vmToUpdate, vmStatistics);
Line 1980: 
Line 1981:             try {
> This is where we use the diskImages we loaded earlier
i referred to the try-catch clause
Line 1982:                 for (DiskImageDynamic imageDynamic : 
_runningVms.get(vmToUpdate.getId()).getVmDynamic().getDisks()) {
Line 1983:                     Guid activeImageId = 
diskImages.get(imageDynamic.getId());
Line 1984:                     // We have disk_id statistics, which is good, 
but disk_image_dynamic table contains image_id, so we
Line 1985:                     // update for the AI.


Line 1980: 
Line 1981:             try {
Line 1982:                 for (DiskImageDynamic imageDynamic : 
_runningVms.get(vmToUpdate.getId()).getVmDynamic().getDisks()) {
Line 1983:                     Guid activeImageId = 
diskImages.get(imageDynamic.getId());
Line 1984:                     // We have disk_id statistics, which is good, 
but disk_image_dynamic table contains image_id, so we
> If we use n as the number of updates, and m as the number of reads, we will
that way we save that unnecessary loads the load is here just to perform the 
update, so we can easily encapsulate it within the update stored procedure 
which will improve things even more imo (less code added here, no need to add 
dedicated stored procedure and better performance). I think that this approach 
has benefits, I'd suggest to reconsider it.
Line 1985:                     // update for the AI.
Line 1986:                     // We also check if the disk is null, as, for 
external VMs the disk is not in the database
Line 1987:                     if (activeImageId != null) {
Line 1988:                         imageDynamic.setId(activeImageId);


-- 
To view, visit http://gerrit.ovirt.org/27586
To unsubscribe, visit http://gerrit.ovirt.org/settings

Gerrit-MessageType: comment
Gerrit-Change-Id: I94e20d782bc4e2befaf4338f51551a2855509769
Gerrit-PatchSet: 10
Gerrit-Project: ovirt-engine
Gerrit-Branch: master
Gerrit-Owner: Liran Zelkha <lzel...@redhat.com>
Gerrit-Reviewer: Allon Mureinik <amure...@redhat.com>
Gerrit-Reviewer: Barak Azulay <bazu...@redhat.com>
Gerrit-Reviewer: Eli Mesika <emes...@redhat.com>
Gerrit-Reviewer: Liran Zelkha <lzel...@redhat.com>
Gerrit-Reviewer: Liron Ar <lara...@redhat.com>
Gerrit-Reviewer: Omer Frenkel <ofren...@redhat.com>
Gerrit-Reviewer: Oved Ourfali <oourf...@redhat.com>
Gerrit-Reviewer: Roy Golan <rgo...@redhat.com>
Gerrit-Reviewer: automat...@ovirt.org
Gerrit-Reviewer: oVirt Jenkins CI Server
Gerrit-HasComments: Yes
_______________________________________________
Engine-patches mailing list
Engine-patches@ovirt.org
http://lists.ovirt.org/mailman/listinfo/engine-patches

Reply via email to