[jira] [Updated] (MAPREDUCE-4228) mapreduce.job.reduce.slowstart.completedmaps is not working properly to delay the scheduling of the reduce tasks
[ https://issues.apache.org/jira/browse/MAPREDUCE-4228?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Robert Joseph Evans updated MAPREDUCE-4228: --- Resolution: Fixed Fix Version/s: 3.0.0 2.0.1-alpha 0.23.3 Status: Resolved (was: Patch Available) Thanks for the patch Jason, I have put this into trunk, branch-2, and branch-0.23. > mapreduce.job.reduce.slowstart.completedmaps is not working properly to delay > the scheduling of the reduce tasks > > > Key: MAPREDUCE-4228 > URL: https://issues.apache.org/jira/browse/MAPREDUCE-4228 > Project: Hadoop Map/Reduce > Issue Type: Bug > Components: applicationmaster, mrv2 >Affects Versions: 0.23.1 >Reporter: Jason Lowe >Assignee: Jason Lowe > Fix For: 0.23.3, 2.0.1-alpha, 3.0.0 > > Attachments: MAPREDUCE-4228.patch, MAPREDUCE-4228.patch, > MAPREDUCE-4228.patch > > > If no more map tasks need to be scheduled but not all have completed, the > ApplicationMaster will start scheduling reducers even if the number of > completed maps has not met the mapreduce.job.reduce.slowstart.completedmaps > threshold. For example, if the property is set to 1.0 all maps should > complete before any reducers are scheduled. However the reducers are > scheduled as soon as the last map task is assigned to a container. For a job > with very long-running maps, a cluster with enough capacity to launch all map > tasks could cause reducers to launch prematurely and waste cluster resources. > Thanks to Phil Su for discovering this issue. -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Updated] (MAPREDUCE-4228) mapreduce.job.reduce.slowstart.completedmaps is not working properly to delay the scheduling of the reduce tasks
[ https://issues.apache.org/jira/browse/MAPREDUCE-4228?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Jason Lowe updated MAPREDUCE-4228: -- Attachment: MAPREDUCE-4228.patch Upmerged patch to latest changes on trunk. > mapreduce.job.reduce.slowstart.completedmaps is not working properly to delay > the scheduling of the reduce tasks > > > Key: MAPREDUCE-4228 > URL: https://issues.apache.org/jira/browse/MAPREDUCE-4228 > Project: Hadoop Map/Reduce > Issue Type: Bug > Components: applicationmaster, mrv2 >Affects Versions: 0.23.1 >Reporter: Jason Lowe >Assignee: Jason Lowe > Attachments: MAPREDUCE-4228.patch, MAPREDUCE-4228.patch, > MAPREDUCE-4228.patch > > > If no more map tasks need to be scheduled but not all have completed, the > ApplicationMaster will start scheduling reducers even if the number of > completed maps has not met the mapreduce.job.reduce.slowstart.completedmaps > threshold. For example, if the property is set to 1.0 all maps should > complete before any reducers are scheduled. However the reducers are > scheduled as soon as the last map task is assigned to a container. For a job > with very long-running maps, a cluster with enough capacity to launch all map > tasks could cause reducers to launch prematurely and waste cluster resources. > Thanks to Phil Su for discovering this issue. -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Updated] (MAPREDUCE-4228) mapreduce.job.reduce.slowstart.completedmaps is not working properly to delay the scheduling of the reduce tasks
[ https://issues.apache.org/jira/browse/MAPREDUCE-4228?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Jason Lowe updated MAPREDUCE-4228: -- Attachment: MAPREDUCE-4228.patch Minor patch update to add Private annotation on scheduleAllReduces() which was made public for testing. > mapreduce.job.reduce.slowstart.completedmaps is not working properly to delay > the scheduling of the reduce tasks > > > Key: MAPREDUCE-4228 > URL: https://issues.apache.org/jira/browse/MAPREDUCE-4228 > Project: Hadoop Map/Reduce > Issue Type: Bug > Components: applicationmaster, mrv2 >Affects Versions: 0.23.1 >Reporter: Jason Lowe >Assignee: Jason Lowe > Attachments: MAPREDUCE-4228.patch, MAPREDUCE-4228.patch > > > If no more map tasks need to be scheduled but not all have completed, the > ApplicationMaster will start scheduling reducers even if the number of > completed maps has not met the mapreduce.job.reduce.slowstart.completedmaps > threshold. For example, if the property is set to 1.0 all maps should > complete before any reducers are scheduled. However the reducers are > scheduled as soon as the last map task is assigned to a container. For a job > with very long-running maps, a cluster with enough capacity to launch all map > tasks could cause reducers to launch prematurely and waste cluster resources. > Thanks to Phil Su for discovering this issue. -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Updated] (MAPREDUCE-4228) mapreduce.job.reduce.slowstart.completedmaps is not working properly to delay the scheduling of the reduce tasks
[ https://issues.apache.org/jira/browse/MAPREDUCE-4228?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Jason Lowe updated MAPREDUCE-4228: -- Status: Patch Available (was: Open) > mapreduce.job.reduce.slowstart.completedmaps is not working properly to delay > the scheduling of the reduce tasks > > > Key: MAPREDUCE-4228 > URL: https://issues.apache.org/jira/browse/MAPREDUCE-4228 > Project: Hadoop Map/Reduce > Issue Type: Bug > Components: applicationmaster, mrv2 >Affects Versions: 0.23.1 >Reporter: Jason Lowe >Assignee: Jason Lowe > Attachments: MAPREDUCE-4228.patch, MAPREDUCE-4228.patch > > > If no more map tasks need to be scheduled but not all have completed, the > ApplicationMaster will start scheduling reducers even if the number of > completed maps has not met the mapreduce.job.reduce.slowstart.completedmaps > threshold. For example, if the property is set to 1.0 all maps should > complete before any reducers are scheduled. However the reducers are > scheduled as soon as the last map task is assigned to a container. For a job > with very long-running maps, a cluster with enough capacity to launch all map > tasks could cause reducers to launch prematurely and waste cluster resources. > Thanks to Phil Su for discovering this issue. -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Updated] (MAPREDUCE-4228) mapreduce.job.reduce.slowstart.completedmaps is not working properly to delay the scheduling of the reduce tasks
[ https://issues.apache.org/jira/browse/MAPREDUCE-4228?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Jason Lowe updated MAPREDUCE-4228: -- Status: Open (was: Patch Available) > mapreduce.job.reduce.slowstart.completedmaps is not working properly to delay > the scheduling of the reduce tasks > > > Key: MAPREDUCE-4228 > URL: https://issues.apache.org/jira/browse/MAPREDUCE-4228 > Project: Hadoop Map/Reduce > Issue Type: Bug > Components: applicationmaster, mrv2 >Affects Versions: 0.23.1 >Reporter: Jason Lowe >Assignee: Jason Lowe > Attachments: MAPREDUCE-4228.patch > > > If no more map tasks need to be scheduled but not all have completed, the > ApplicationMaster will start scheduling reducers even if the number of > completed maps has not met the mapreduce.job.reduce.slowstart.completedmaps > threshold. For example, if the property is set to 1.0 all maps should > complete before any reducers are scheduled. However the reducers are > scheduled as soon as the last map task is assigned to a container. For a job > with very long-running maps, a cluster with enough capacity to launch all map > tasks could cause reducers to launch prematurely and waste cluster resources. > Thanks to Phil Su for discovering this issue. -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Updated] (MAPREDUCE-4228) mapreduce.job.reduce.slowstart.completedmaps is not working properly to delay the scheduling of the reduce tasks
[ https://issues.apache.org/jira/browse/MAPREDUCE-4228?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Jason Lowe updated MAPREDUCE-4228: -- Target Version/s: 0.23.3 Status: Patch Available (was: Open) > mapreduce.job.reduce.slowstart.completedmaps is not working properly to delay > the scheduling of the reduce tasks > > > Key: MAPREDUCE-4228 > URL: https://issues.apache.org/jira/browse/MAPREDUCE-4228 > Project: Hadoop Map/Reduce > Issue Type: Bug > Components: applicationmaster, mrv2 >Affects Versions: 0.23.1 >Reporter: Jason Lowe >Assignee: Jason Lowe > Attachments: MAPREDUCE-4228.patch > > > If no more map tasks need to be scheduled but not all have completed, the > ApplicationMaster will start scheduling reducers even if the number of > completed maps has not met the mapreduce.job.reduce.slowstart.completedmaps > threshold. For example, if the property is set to 1.0 all maps should > complete before any reducers are scheduled. However the reducers are > scheduled as soon as the last map task is assigned to a container. For a job > with very long-running maps, a cluster with enough capacity to launch all map > tasks could cause reducers to launch prematurely and waste cluster resources. > Thanks to Phil Su for discovering this issue. -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Updated] (MAPREDUCE-4228) mapreduce.job.reduce.slowstart.completedmaps is not working properly to delay the scheduling of the reduce tasks
[ https://issues.apache.org/jira/browse/MAPREDUCE-4228?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Jason Lowe updated MAPREDUCE-4228: -- Attachment: MAPREDUCE-4228.patch Patch to verify we've met the slowstart criteria before unleashing all reducers once all maps have been assigned. Besides the automated tests that were added/modified, I also ran manual tests on a local cluster with sleep and wordcount jobs. > mapreduce.job.reduce.slowstart.completedmaps is not working properly to delay > the scheduling of the reduce tasks > > > Key: MAPREDUCE-4228 > URL: https://issues.apache.org/jira/browse/MAPREDUCE-4228 > Project: Hadoop Map/Reduce > Issue Type: Bug > Components: applicationmaster, mrv2 >Affects Versions: 0.23.1 >Reporter: Jason Lowe >Assignee: Jason Lowe > Attachments: MAPREDUCE-4228.patch > > > If no more map tasks need to be scheduled but not all have completed, the > ApplicationMaster will start scheduling reducers even if the number of > completed maps has not met the mapreduce.job.reduce.slowstart.completedmaps > threshold. For example, if the property is set to 1.0 all maps should > complete before any reducers are scheduled. However the reducers are > scheduled as soon as the last map task is assigned to a container. For a job > with very long-running maps, a cluster with enough capacity to launch all map > tasks could cause reducers to launch prematurely and waste cluster resources. > Thanks to Phil Su for discovering this issue. -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa For more information on JIRA, see: http://www.atlassian.com/software/jira
