[jira] [Updated] (MAPREDUCE-4502) Multi-level aggregation with combining the result of maps per node/rack
[ https://issues.apache.org/jira/browse/MAPREDUCE-4502?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Tsuyoshi OZAWA updated MAPREDUCE-4502: -- Attachment: design_v3.pdf Updated design note including benchmark of node-level aggregation. This revision dedicates only to explain node-level aggregation design and improvement of performance. The rack-level aggregation still needs big change, so I'd like to discuss in the other ticket. > Multi-level aggregation with combining the result of maps per node/rack > --- > > Key: MAPREDUCE-4502 > URL: https://issues.apache.org/jira/browse/MAPREDUCE-4502 > Project: Hadoop Map/Reduce > Issue Type: Improvement > Components: applicationmaster, mrv2 >Affects Versions: 3.0.0 >Reporter: Tsuyoshi OZAWA >Assignee: Tsuyoshi OZAWA > Attachments: design_v2.pdf, design_v3.pdf, MAPREDUCE-4502.1.patch, > MAPREDUCE-4502.2.patch, MAPREDUCE-4502.3.patch, MAPREDUCE-4502.4.patch, > MAPREDUCE-4502.5.patch, MAPREDUCE-4502.6.patch, MAPREDUCE-4525-pof.diff, > speculative_draft.pdf > > > The shuffle costs is expensive in Hadoop in spite of the existence of > combiner, because the scope of combining is limited within only one MapTask. > To solve this problem, it's a good way to aggregate the result of maps per > node/rack by launch combiner. > This JIRA is to implement the multi-level aggregation infrastructure, > including combining per container(MAPREDUCE-3902 is related), coordinating > containers by application master without breaking fault tolerance of jobs. -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Updated] (MAPREDUCE-4502) Multi-level aggregation with combining the result of maps per node/rack
[ https://issues.apache.org/jira/browse/MAPREDUCE-4502?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Tsuyoshi OZAWA updated MAPREDUCE-4502: -- Attachment: MAPREDUCE-4502.6.patch Oops, I attached the wrong patch. This is correct one. > Multi-level aggregation with combining the result of maps per node/rack > --- > > Key: MAPREDUCE-4502 > URL: https://issues.apache.org/jira/browse/MAPREDUCE-4502 > Project: Hadoop Map/Reduce > Issue Type: Improvement > Components: applicationmaster, mrv2 >Affects Versions: 3.0.0 >Reporter: Tsuyoshi OZAWA >Assignee: Tsuyoshi OZAWA > Attachments: design_v2.pdf, MAPREDUCE-4502.1.patch, > MAPREDUCE-4502.2.patch, MAPREDUCE-4502.3.patch, MAPREDUCE-4502.4.patch, > MAPREDUCE-4502.5.patch, MAPREDUCE-4502.6.patch, MAPREDUCE-4525-pof.diff, > speculative_draft.pdf > > > The shuffle costs is expensive in Hadoop in spite of the existence of > combiner, because the scope of combining is limited within only one MapTask. > To solve this problem, it's a good way to aggregate the result of maps per > node/rack by launch combiner. > This JIRA is to implement the multi-level aggregation infrastructure, > including combining per container(MAPREDUCE-3902 is related), coordinating > containers by application master without breaking fault tolerance of jobs. -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Updated] (MAPREDUCE-4502) Multi-level aggregation with combining the result of maps per node/rack
[ https://issues.apache.org/jira/browse/MAPREDUCE-4502?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Tsuyoshi OZAWA updated MAPREDUCE-4502: -- Attachment: MAPREDUCE-4502.5.patch HADOOP-9112 caused regression, so I retry it based on today's trunk. > Multi-level aggregation with combining the result of maps per node/rack > --- > > Key: MAPREDUCE-4502 > URL: https://issues.apache.org/jira/browse/MAPREDUCE-4502 > Project: Hadoop Map/Reduce > Issue Type: Improvement > Components: applicationmaster, mrv2 >Affects Versions: 3.0.0 >Reporter: Tsuyoshi OZAWA >Assignee: Tsuyoshi OZAWA > Attachments: design_v2.pdf, MAPREDUCE-4502.1.patch, > MAPREDUCE-4502.2.patch, MAPREDUCE-4502.3.patch, MAPREDUCE-4502.4.patch, > MAPREDUCE-4502.5.patch, MAPREDUCE-4525-pof.diff, speculative_draft.pdf > > > The shuffle costs is expensive in Hadoop in spite of the existence of > combiner, because the scope of combining is limited within only one MapTask. > To solve this problem, it's a good way to aggregate the result of maps per > node/rack by launch combiner. > This JIRA is to implement the multi-level aggregation infrastructure, > including combining per container(MAPREDUCE-3902 is related), coordinating > containers by application master without breaking fault tolerance of jobs. -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Updated] (MAPREDUCE-4502) Multi-level aggregation with combining the result of maps per node/rack
[ https://issues.apache.org/jira/browse/MAPREDUCE-4502?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Tsuyoshi OZAWA updated MAPREDUCE-4502: -- Attachment: MAPREDUCE-4502.4.patch Fixed to handle exception correctly and add timeout to TestJobConf. > Multi-level aggregation with combining the result of maps per node/rack > --- > > Key: MAPREDUCE-4502 > URL: https://issues.apache.org/jira/browse/MAPREDUCE-4502 > Project: Hadoop Map/Reduce > Issue Type: Improvement > Components: applicationmaster, mrv2 >Affects Versions: 3.0.0 >Reporter: Tsuyoshi OZAWA >Assignee: Tsuyoshi OZAWA > Attachments: design_v2.pdf, MAPREDUCE-4502.1.patch, > MAPREDUCE-4502.2.patch, MAPREDUCE-4502.3.patch, MAPREDUCE-4502.4.patch, > MAPREDUCE-4525-pof.diff, speculative_draft.pdf > > > The shuffle costs is expensive in Hadoop in spite of the existence of > combiner, because the scope of combining is limited within only one MapTask. > To solve this problem, it's a good way to aggregate the result of maps per > node/rack by launch combiner. > This JIRA is to implement the multi-level aggregation infrastructure, > including combining per container(MAPREDUCE-3902 is related), coordinating > containers by application master without breaking fault tolerance of jobs. -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Updated] (MAPREDUCE-4502) Multi-level aggregation with combining the result of maps per node/rack
[ https://issues.apache.org/jira/browse/MAPREDUCE-4502?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Tsuyoshi OZAWA updated MAPREDUCE-4502: -- Attachment: MAPREDUCE-4502.3.patch Fixed the overlooked findbugs warnings and Added timeout value to test. > Multi-level aggregation with combining the result of maps per node/rack > --- > > Key: MAPREDUCE-4502 > URL: https://issues.apache.org/jira/browse/MAPREDUCE-4502 > Project: Hadoop Map/Reduce > Issue Type: Improvement > Components: applicationmaster, mrv2 >Affects Versions: 3.0.0 >Reporter: Tsuyoshi OZAWA >Assignee: Tsuyoshi OZAWA > Attachments: design_v2.pdf, MAPREDUCE-4502.1.patch, > MAPREDUCE-4502.2.patch, MAPREDUCE-4502.3.patch, MAPREDUCE-4525-pof.diff, > speculative_draft.pdf > > > The shuffle costs is expensive in Hadoop in spite of the existence of > combiner, because the scope of combining is limited within only one MapTask. > To solve this problem, it's a good way to aggregate the result of maps per > node/rack by launch combiner. > This JIRA is to implement the multi-level aggregation infrastructure, > including combining per container(MAPREDUCE-3902 is related), coordinating > containers by application master without breaking fault tolerance of jobs. -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Updated] (MAPREDUCE-4502) Multi-level aggregation with combining the result of maps per node/rack
[ https://issues.apache.org/jira/browse/MAPREDUCE-4502?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Tsuyoshi OZAWA updated MAPREDUCE-4502: -- Attachment: MAPREDUCE-4502.2.patch Fixed findbugs and release audit warnings. > Multi-level aggregation with combining the result of maps per node/rack > --- > > Key: MAPREDUCE-4502 > URL: https://issues.apache.org/jira/browse/MAPREDUCE-4502 > Project: Hadoop Map/Reduce > Issue Type: Improvement > Components: applicationmaster, mrv2 >Affects Versions: 3.0.0 >Reporter: Tsuyoshi OZAWA >Assignee: Tsuyoshi OZAWA > Attachments: design_v2.pdf, MAPREDUCE-4502.1.patch, > MAPREDUCE-4502.2.patch, MAPREDUCE-4525-pof.diff, speculative_draft.pdf > > > The shuffle costs is expensive in Hadoop in spite of the existence of > combiner, because the scope of combining is limited within only one MapTask. > To solve this problem, it's a good way to aggregate the result of maps per > node/rack by launch combiner. > This JIRA is to implement the multi-level aggregation infrastructure, > including combining per container(MAPREDUCE-3902 is related), coordinating > containers by application master without breaking fault tolerance of jobs. -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Updated] (MAPREDUCE-4502) Multi-level aggregation with combining the result of maps per node/rack
[ https://issues.apache.org/jira/browse/MAPREDUCE-4502?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Tsuyoshi OZAWA updated MAPREDUCE-4502: -- Affects Version/s: 3.0.0 Status: Patch Available (was: Open) > Multi-level aggregation with combining the result of maps per node/rack > --- > > Key: MAPREDUCE-4502 > URL: https://issues.apache.org/jira/browse/MAPREDUCE-4502 > Project: Hadoop Map/Reduce > Issue Type: Improvement > Components: applicationmaster, mrv2 >Affects Versions: 3.0.0 >Reporter: Tsuyoshi OZAWA >Assignee: Tsuyoshi OZAWA > Attachments: design_v2.pdf, MAPREDUCE-4502.1.patch, > MAPREDUCE-4525-pof.diff, speculative_draft.pdf > > > The shuffle costs is expensive in Hadoop in spite of the existence of > combiner, because the scope of combining is limited within only one MapTask. > To solve this problem, it's a good way to aggregate the result of maps per > node/rack by launch combiner. > This JIRA is to implement the multi-level aggregation infrastructure, > including combining per container(MAPREDUCE-3902 is related), coordinating > containers by application master without breaking fault tolerance of jobs. -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Updated] (MAPREDUCE-4502) Multi-level aggregation with combining the result of maps per node/rack
[ https://issues.apache.org/jira/browse/MAPREDUCE-4502?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Tsuyoshi OZAWA updated MAPREDUCE-4502: -- Attachment: MAPREDUCE-4502.1.patch Added patch including test. > Multi-level aggregation with combining the result of maps per node/rack > --- > > Key: MAPREDUCE-4502 > URL: https://issues.apache.org/jira/browse/MAPREDUCE-4502 > Project: Hadoop Map/Reduce > Issue Type: Improvement > Components: applicationmaster, mrv2 >Reporter: Tsuyoshi OZAWA >Assignee: Tsuyoshi OZAWA > Attachments: design_v2.pdf, MAPREDUCE-4502.1.patch, > MAPREDUCE-4525-pof.diff, speculative_draft.pdf > > > The shuffle costs is expensive in Hadoop in spite of the existence of > combiner, because the scope of combining is limited within only one MapTask. > To solve this problem, it's a good way to aggregate the result of maps per > node/rack by launch combiner. > This JIRA is to implement the multi-level aggregation infrastructure, > including combining per container(MAPREDUCE-3902 is related), coordinating > containers by application master without breaking fault tolerance of jobs. -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Updated] (MAPREDUCE-4502) Multi-level aggregation with combining the result of maps per node/rack
[ https://issues.apache.org/jira/browse/MAPREDUCE-4502?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Tsuyoshi OZAWA updated MAPREDUCE-4502: -- Attachment: MAPREDUCE-4525-pof.diff I've prototyped proof of concept code of node-level aggregation. I made sure that it works well, so I share it! Current situation is as follows: * Node level aggregation in distributed mode is supported. I made sure of the correctness of the result by using diff command. * Fault tolerance is partially supported. At design level, it can be supported(WIP). * No test codes. * No switch on/off feature. Quick benchmark is as follows: |Default Hadoop(sec)|Node level aggregation(sec)| |105.0|82.0| Experiment Environment is as follows: * Input file is 1GB, generated by RandomTextWriter. * 8 CPU/machines. * 2 machines. I'm appreciate if you review the design and code. And, I have several questions to deal with: 1. How can I write tests such a complex case? Any ideas or references? 2. I've extended mr_protos.proto. Is it acceptable? My next step is creating patch sets with tests and fully fault tolerance support to merge it into trunk. I'm also planning to benchmark of node-level aggregation. I'll show you here when I do it. Note that design note is outdated, so I'm updating it now. If you'd like to track changes my code changes, you can see it on [my github|https://github.com/oza/hadoop-common/tree/MAPREDUCE-4525-Node-level-aggregation]. > Multi-level aggregation with combining the result of maps per node/rack > --- > > Key: MAPREDUCE-4502 > URL: https://issues.apache.org/jira/browse/MAPREDUCE-4502 > Project: Hadoop Map/Reduce > Issue Type: Improvement > Components: applicationmaster, mrv2 >Reporter: Tsuyoshi OZAWA >Assignee: Tsuyoshi OZAWA > Attachments: design_v2.pdf, MAPREDUCE-4525-pof.diff, > speculative_draft.pdf > > > The shuffle costs is expensive in Hadoop in spite of the existence of > combiner, because the scope of combining is limited within only one MapTask. > To solve this problem, it's a good way to aggregate the result of maps per > node/rack by launch combiner. > This JIRA is to implement the multi-level aggregation infrastructure, > including combining per container(MAPREDUCE-3902 is related), coordinating > containers by application master without breaking fault tolerance of jobs. -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Updated] (MAPREDUCE-4502) Multi-level aggregation with combining the result of maps per node/rack
[ https://issues.apache.org/jira/browse/MAPREDUCE-4502?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Tsuyoshi OZAWA updated MAPREDUCE-4502: -- Attachment: design_v2.pdf I redesigned its architecture with the ideas inspired by discussion among Chris and Karthik. Main diffs are as follows: 1. All aggregations are preformed in containers(at the end of MapTasks). Because of this change, the umbilical protocol is need to change. 3. No task types are added newly. 4. Rack-aware aggregation is taken into account. Any feedback are welcome. Thanks, - Tsuyoshi > Multi-level aggregation with combining the result of maps per node/rack > --- > > Key: MAPREDUCE-4502 > URL: https://issues.apache.org/jira/browse/MAPREDUCE-4502 > Project: Hadoop Map/Reduce > Issue Type: Improvement > Components: applicationmaster, mrv2 >Reporter: Tsuyoshi OZAWA >Assignee: Tsuyoshi OZAWA > Attachments: design_v2.pdf, speculative_draft.pdf > > > The shuffle costs is expensive in Hadoop in spite of the existence of > combiner, because the scope of combining is limited within only one MapTask. > To solve this problem, it's a good way to aggregate the result of maps per > node/rack by launch combiner. > This JIRA is to implement the multi-level aggregation infrastructure, > including combining per container(MAPREDUCE-3902 is related), coordinating > containers by application master without breaking fault tolerance of jobs. -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Updated] (MAPREDUCE-4502) Multi-level aggregation with combining the result of maps per node/rack
[ https://issues.apache.org/jira/browse/MAPREDUCE-4502?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Tsuyoshi OZAWA updated MAPREDUCE-4502: -- Attachment: speculative_draft.pdf I attached the first speculative draft design of multi-level aggregation. My main concern is whether the changes described in this document are acceptable or not for Hadoop community. Please let me know if you have any opinion. Thanks! > Multi-level aggregation with combining the result of maps per node/rack > --- > > Key: MAPREDUCE-4502 > URL: https://issues.apache.org/jira/browse/MAPREDUCE-4502 > Project: Hadoop Map/Reduce > Issue Type: Improvement > Components: applicationmaster, mrv2 >Reporter: Tsuyoshi OZAWA >Assignee: Tsuyoshi OZAWA > Attachments: speculative_draft.pdf > > > The shuffle costs is expensive in Hadoop in spite of the existence of > combiner, because the scope of combining is limited within only one MapTask. > To solve this problem, it's a good way to aggregate the result of maps per > node/rack by launch combiner. > This JIRA is to implement the multi-level aggregation infrastructure, > including combining per container(MAPREDUCE-3902 is related), coordinating > containers by application master without breaking fault tolerance of jobs. -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Updated] (MAPREDUCE-4502) Multi-level aggregation with combining the result of maps per node/rack
[ https://issues.apache.org/jira/browse/MAPREDUCE-4502?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Tsuyoshi OZAWA updated MAPREDUCE-4502: -- Assignee: Tsuyoshi OZAWA > Multi-level aggregation with combining the result of maps per node/rack > --- > > Key: MAPREDUCE-4502 > URL: https://issues.apache.org/jira/browse/MAPREDUCE-4502 > Project: Hadoop Map/Reduce > Issue Type: Improvement > Components: applicationmaster, mrv2 >Reporter: Tsuyoshi OZAWA >Assignee: Tsuyoshi OZAWA > > The shuffle costs is expensive in Hadoop in spite of the existence of > combiner, because the scope of combining is limited within only one MapTask. > To solve this problem, it's a good way to aggregate the result of maps per > node/rack by launch combiner. > This JIRA is to implement the multi-level aggregation infrastructure, > including combining per container(MAPREDUCE-3902 is related), coordinating > containers by application master without breaking fault tolerance of jobs. -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Updated] (MAPREDUCE-4502) Multi-level aggregation with combining the result of maps per node/rack
[ https://issues.apache.org/jira/browse/MAPREDUCE-4502?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Tsuyoshi OZAWA updated MAPREDUCE-4502: -- Description: The shuffle costs is expensive in Hadoop in spite of the existence of combiner, because the scope of combining is limited within only one MapTask. To solve this problem, it's a good way to aggregate the result of maps per node/rack by launch combiner. This JIRA is to implement the multi-level aggregation infrastructure, including combining per container(MAPREDUCE-3902 is related), coordinating containers by application master without breaking fault tolerance of jobs. was: The shuffle costs is expensive in Hadoop in spite of the existence of combiner, because the scope of combining is limited within only one MapTask. To solve this problem, it's a good way to aggregate the result of maps per node/rack by launch combiner. This JIRA is to implement the multi-level aggregation infrastructure, including combining per container(MAPREDUCE-3902 is related), coordinating containers by application master without breaking fault tolerance of jobs. > Multi-level aggregation with combining the result of maps per node/rack > --- > > Key: MAPREDUCE-4502 > URL: https://issues.apache.org/jira/browse/MAPREDUCE-4502 > Project: Hadoop Map/Reduce > Issue Type: Improvement > Components: applicationmaster, mrv2 >Reporter: Tsuyoshi OZAWA > > The shuffle costs is expensive in Hadoop in spite of the existence of > combiner, because the scope of combining is limited within only one MapTask. > To solve this problem, it's a good way to aggregate the result of maps per > node/rack by launch combiner. > This JIRA is to implement the multi-level aggregation infrastructure, > including combining per container(MAPREDUCE-3902 is related), coordinating > containers by application master without breaking fault tolerance of jobs. -- 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