[jira] [Updated] (YARN-2004) Priority scheduling support in Capacity scheduler
[
https://issues.apache.org/jira/browse/YARN-2004?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Sunil G updated YARN-2004:
--
Attachment: 0010-YARN-2004.patch
Hi [~leftnoteasy]
Thank you very much for sharing the comments. I have updated the patch
addressing the comments.
- applicationComparator is still kept here. I raised a ticket to remove it,
once that is done, i will rebase this patch.
- {{FairScheduler#getAppWeight}} I understood the idea very much. I feel we can
have this later as an improvement once the base version is done. How do u feel?
Priority scheduling support in Capacity scheduler
-
Key: YARN-2004
URL: https://issues.apache.org/jira/browse/YARN-2004
Project: Hadoop YARN
Issue Type: Sub-task
Components: capacityscheduler
Reporter: Sunil G
Assignee: Sunil G
Attachments: 0001-YARN-2004.patch, 0002-YARN-2004.patch,
0003-YARN-2004.patch, 0004-YARN-2004.patch, 0005-YARN-2004.patch,
0006-YARN-2004.patch, 0007-YARN-2004.patch, 0008-YARN-2004.patch,
0009-YARN-2004.patch, 0010-YARN-2004.patch
Based on the priority of the application, Capacity Scheduler should be able
to give preference to application while doing scheduling.
ComparatorFiCaSchedulerApp applicationComparator can be changed as below.
1.Check for Application priority. If priority is available, then return
the highest priority job.
2.Otherwise continue with existing logic such as App ID comparison and
then TimeStamp comparison.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
[jira] [Updated] (YARN-2004) Priority scheduling support in Capacity scheduler
[
https://issues.apache.org/jira/browse/YARN-2004?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Sunil G updated YARN-2004:
--
Attachment: 0009-YARN-2004.patch
Thank you very much [~jianhe] and [~leftnoteasy] for the review and comments
- bq.The app priority seems only used for pending applications, how about
priority support for the actively running applications ?
As mentioned in Wangda's comments, I was using application priority comparision
in {{SchedulerApplicationAttempt#compareInputOrderTo}}. However its better to
do the same in {{FifoComparator}}, and I have updated SchedulableEntity
interface to get priroity for same.
- bq.Should we do priority comparison first (treat priority as class) OR
combination of them (treat priority as factor).
I feel we can do the priority comparision first. Do you see any specific
usecase for priority as factor
- bq.We may just move the check into RMAppManager.
Adding further to Wangda's points, We had a PriorityManager earlier to store
the same. Since we associated the priroity closer to Scehduler, it is now
handled here.
- bq.updateApplicationPriority
In this patch I have handled this scenario in RMAppImpl (By handling a new
event from Scheduler and storing in StateStore) and CS. RMAppImpl changes are
updated in YARN-2003. Please suggest whether I need to handle this specific
scenario in a separate ticket (RMAppImpl and CS together).
I have fixed all the other comments as mentioned. Kindly help to check.
Priority scheduling support in Capacity scheduler
-
Key: YARN-2004
URL: https://issues.apache.org/jira/browse/YARN-2004
Project: Hadoop YARN
Issue Type: Sub-task
Components: capacityscheduler
Reporter: Sunil G
Assignee: Sunil G
Attachments: 0001-YARN-2004.patch, 0002-YARN-2004.patch,
0003-YARN-2004.patch, 0004-YARN-2004.patch, 0005-YARN-2004.patch,
0006-YARN-2004.patch, 0007-YARN-2004.patch, 0008-YARN-2004.patch,
0009-YARN-2004.patch
Based on the priority of the application, Capacity Scheduler should be able
to give preference to application while doing scheduling.
ComparatorFiCaSchedulerApp applicationComparator can be changed as below.
1.Check for Application priority. If priority is available, then return
the highest priority job.
2.Otherwise continue with existing logic such as App ID comparison and
then TimeStamp comparison.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
[jira] [Updated] (YARN-2004) Priority scheduling support in Capacity scheduler
[
https://issues.apache.org/jira/browse/YARN-2004?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Sunil G updated YARN-2004:
--
Attachment: 0008-YARN-2004.patch
Thank you [~eepayne] for sharing the comments.
bq. Can getApplicationPriority return null?
In CS addApplicationAttempt(), immediately after creating FicaSchedulerApp, I
am setting the priority.
{code}
application.setCurrentAppAttempt(attempt);
attempt.setApplicationPriority(application.getPriority());
{code}
Also from YARN-2003, we set the application priority in all cases while
processing submitApplication. If user didnt set a priority, application will
have the default priority from Queue. If user didnt configure a default
priority in Queue, then by default 0(ZERO) will be considered as default
priority.
Please let me know if this is fine.
Priority scheduling support in Capacity scheduler
-
Key: YARN-2004
URL: https://issues.apache.org/jira/browse/YARN-2004
Project: Hadoop YARN
Issue Type: Sub-task
Components: capacityscheduler
Reporter: Sunil G
Assignee: Sunil G
Attachments: 0001-YARN-2004.patch, 0002-YARN-2004.patch,
0003-YARN-2004.patch, 0004-YARN-2004.patch, 0005-YARN-2004.patch,
0006-YARN-2004.patch, 0007-YARN-2004.patch, 0008-YARN-2004.patch
Based on the priority of the application, Capacity Scheduler should be able
to give preference to application while doing scheduling.
ComparatorFiCaSchedulerApp applicationComparator can be changed as below.
1.Check for Application priority. If priority is available, then return
the highest priority job.
2.Otherwise continue with existing logic such as App ID comparison and
then TimeStamp comparison.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
[jira] [Updated] (YARN-2004) Priority scheduling support in Capacity scheduler
[ https://issues.apache.org/jira/browse/YARN-2004?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Sunil G updated YARN-2004: -- Attachment: 0007-YARN-2004.patch Rebasing patch against latest trunk. Also made changes as per OrderingPolicy in CS. Priority scheduling support in Capacity scheduler - Key: YARN-2004 URL: https://issues.apache.org/jira/browse/YARN-2004 Project: Hadoop YARN Issue Type: Sub-task Components: capacityscheduler Reporter: Sunil G Assignee: Sunil G Attachments: 0001-YARN-2004.patch, 0002-YARN-2004.patch, 0003-YARN-2004.patch, 0004-YARN-2004.patch, 0005-YARN-2004.patch, 0006-YARN-2004.patch, 0007-YARN-2004.patch Based on the priority of the application, Capacity Scheduler should be able to give preference to application while doing scheduling. ComparatorFiCaSchedulerApp applicationComparator can be changed as below. 1.Check for Application priority. If priority is available, then return the highest priority job. 2.Otherwise continue with existing logic such as App ID comparison and then TimeStamp comparison. -- This message was sent by Atlassian JIRA (v6.3.4#6332)
[jira] [Updated] (YARN-2004) Priority scheduling support in Capacity scheduler
[
https://issues.apache.org/jira/browse/YARN-2004?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Sunil G updated YARN-2004:
--
Attachment: 0006-YARN-2004.patch
Updated patch as per the comments. [~leftnoteasy] [~jlowe] Please check the same
bq. If a1.getApplicationPriority() returns non-null but
a2.getApplicationPriority() returns null
I have considered default priority scenario where if submitted app does not
gave any priority, then default will be taken. So chances of null here in above
scenario wont happen.
{noformat}
public int compare(FiCaSchedulerApp a1, FiCaSchedulerApp a2) {
if (!a1.getApplicationPriority().equals(a2.getApplicationPriority())) {
return a1.getApplicationPriority().compareTo(
a2.getApplicationPriority());
}
return a1.getApplicationId().compareTo(a2.getApplicationId());
}
{noformat}
Priority scheduling support in Capacity scheduler
-
Key: YARN-2004
URL: https://issues.apache.org/jira/browse/YARN-2004
Project: Hadoop YARN
Issue Type: Sub-task
Components: capacityscheduler
Reporter: Sunil G
Assignee: Sunil G
Attachments: 0001-YARN-2004.patch, 0002-YARN-2004.patch,
0003-YARN-2004.patch, 0004-YARN-2004.patch, 0005-YARN-2004.patch,
0006-YARN-2004.patch
Based on the priority of the application, Capacity Scheduler should be able
to give preference to application while doing scheduling.
ComparatorFiCaSchedulerApp applicationComparator can be changed as below.
1.Check for Application priority. If priority is available, then return
the highest priority job.
2.Otherwise continue with existing logic such as App ID comparison and
then TimeStamp comparison.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
[jira] [Updated] (YARN-2004) Priority scheduling support in Capacity scheduler
[ https://issues.apache.org/jira/browse/YARN-2004?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Sunil G updated YARN-2004: -- Attachment: 0005-YARN-2004.patch Uploading CS changes. Hi [~leftnoteasy] YARN-2004 need to have some changes in CS and LeafQueue. But same methods dummy impl is added in YARN-2003. Hence it will be dependable with YARN-2003, but not opposite. Kindly share your opinion. Priority scheduling support in Capacity scheduler - Key: YARN-2004 URL: https://issues.apache.org/jira/browse/YARN-2004 Project: Hadoop YARN Issue Type: Sub-task Components: capacityscheduler Reporter: Sunil G Assignee: Sunil G Attachments: 0001-YARN-2004.patch, 0002-YARN-2004.patch, 0003-YARN-2004.patch, 0004-YARN-2004.patch, 0005-YARN-2004.patch Based on the priority of the application, Capacity Scheduler should be able to give preference to application while doing scheduling. ComparatorFiCaSchedulerApp applicationComparator can be changed as below. 1.Check for Application priority. If priority is available, then return the highest priority job. 2.Otherwise continue with existing logic such as App ID comparison and then TimeStamp comparison. -- This message was sent by Atlassian JIRA (v6.3.4#6332)
[jira] [Updated] (YARN-2004) Priority scheduling support in Capacity scheduler
[ https://issues.apache.org/jira/browse/YARN-2004?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Sunil G updated YARN-2004: -- Attachment: 0004-YARN-2004.patch Rebasing the patch Priority scheduling support in Capacity scheduler - Key: YARN-2004 URL: https://issues.apache.org/jira/browse/YARN-2004 Project: Hadoop YARN Issue Type: Sub-task Components: capacityscheduler Reporter: Sunil G Assignee: Sunil G Attachments: 0001-YARN-2004.patch, 0002-YARN-2004.patch, 0003-YARN-2004.patch, 0004-YARN-2004.patch Based on the priority of the application, Capacity Scheduler should be able to give preference to application while doing scheduling. ComparatorFiCaSchedulerApp applicationComparator can be changed as below. 1.Check for Application priority. If priority is available, then return the highest priority job. 2.Otherwise continue with existing logic such as App ID comparison and then TimeStamp comparison. -- This message was sent by Atlassian JIRA (v6.3.4#6332)
[jira] [Updated] (YARN-2004) Priority scheduling support in Capacity scheduler
[ https://issues.apache.org/jira/browse/YARN-2004?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Sunil G updated YARN-2004: -- Attachment: 0003-YARN-2004.patch Updating patch with minor variations Priority scheduling support in Capacity scheduler - Key: YARN-2004 URL: https://issues.apache.org/jira/browse/YARN-2004 Project: Hadoop YARN Issue Type: Sub-task Components: capacityscheduler Reporter: Sunil G Assignee: Sunil G Attachments: 0001-YARN-2004.patch, 0002-YARN-2004.patch, 0003-YARN-2004.patch Based on the priority of the application, Capacity Scheduler should be able to give preference to application while doing scheduling. ComparatorFiCaSchedulerApp applicationComparator can be changed as below. 1.Check for Application priority. If priority is available, then return the highest priority job. 2.Otherwise continue with existing logic such as App ID comparison and then TimeStamp comparison. -- This message was sent by Atlassian JIRA (v6.3.4#6332)
[jira] [Updated] (YARN-2004) Priority scheduling support in Capacity scheduler
[ https://issues.apache.org/jira/browse/YARN-2004?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Sunil G updated YARN-2004: -- Attachment: 0002-YARN-2004.patch Priority scheduling support in Capacity scheduler - Key: YARN-2004 URL: https://issues.apache.org/jira/browse/YARN-2004 Project: Hadoop YARN Issue Type: Sub-task Components: capacityscheduler Reporter: Sunil G Assignee: Sunil G Attachments: 0001-YARN-2004.patch, 0002-YARN-2004.patch Based on the priority of the application, Capacity Scheduler should be able to give preference to application while doing scheduling. ComparatorFiCaSchedulerApp applicationComparator can be changed as below. 1.Check for Application priority. If priority is available, then return the highest priority job. 2.Otherwise continue with existing logic such as App ID comparison and then TimeStamp comparison. -- This message was sent by Atlassian JIRA (v6.3.4#6332)
[jira] [Updated] (YARN-2004) Priority scheduling support in Capacity scheduler
[ https://issues.apache.org/jira/browse/YARN-2004?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Sunil G updated YARN-2004: -- Attachment: 0001-YARN-2004.patch Attaching an initial patch for CS. This does not cover headroom level adjustment which is needed for priority inversion problem. It can be handled in a separate ticket. Kindly review the patch. Priority scheduling support in Capacity scheduler - Key: YARN-2004 URL: https://issues.apache.org/jira/browse/YARN-2004 Project: Hadoop YARN Issue Type: Sub-task Components: capacityscheduler Reporter: Sunil G Assignee: Sunil G Attachments: 0001-YARN-2004.patch Based on the priority of the application, Capacity Scheduler should be able to give preference to application while doing scheduling. ComparatorFiCaSchedulerApp applicationComparator can be changed as below. 1.Check for Application priority. If priority is available, then return the highest priority job. 2.Otherwise continue with existing logic such as App ID comparison and then TimeStamp comparison. -- This message was sent by Atlassian JIRA (v6.3.4#6332)
