[jira] Updated: (HIVE-223) when using map-side aggregates - perform single map-reduce group-by
[ https://issues.apache.org/jira/browse/HIVE-223?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Zheng Shao updated HIVE-223: Resolution: Fixed Fix Version/s: 0.3.0 Release Note: HIVE-223. Single map-reduce group-by using map-side aggregation. (Namit Jain via zshao) Hadoop Flags: [Reviewed] Status: Resolved (was: Patch Available) Committed revision 751166. Thanks Namit! > when using map-side aggregates - perform single map-reduce group-by > --- > > Key: HIVE-223 > URL: https://issues.apache.org/jira/browse/HIVE-223 > Project: Hadoop Hive > Issue Type: Improvement > Components: Query Processor >Reporter: Joydeep Sen Sarma >Assignee: Namit Jain > Fix For: 0.3.0 > > Attachments: 223.2.txt, 223.3.txt, 223.patch1.txt, hive.223.4.patch, > patch.txt > > > today even when we do map side aggregates - we do multiple map-reduce jobs. > however - the reason for doing multiple map-reduce group-bys (for single > group-bys) was the fear of skews. When we are doing map side aggregates - > skews should not exist for the most part. There can be two reason for skews: > - large number of entries for a single grouping set - map side aggregates > should take care of this > - badness in hash function that sends too much stuff to one reducer - we > should be able to take care of this by having good hash functions (and prime > number reducer counts) > So i think we should be able to do a single stage map-reduce when doing > map-side aggregates. -- This message is automatically generated by JIRA. - You can reply to this email to add a comment to the issue online.
[jira] Updated: (HIVE-223) when using map-side aggregates - perform single map-reduce group-by
[ https://issues.apache.org/jira/browse/HIVE-223?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Namit Jain updated HIVE-223: Attachment: hive.223.4.patch > when using map-side aggregates - perform single map-reduce group-by > --- > > Key: HIVE-223 > URL: https://issues.apache.org/jira/browse/HIVE-223 > Project: Hadoop Hive > Issue Type: Improvement > Components: Query Processor >Reporter: Joydeep Sen Sarma >Assignee: Namit Jain > Attachments: 223.2.txt, 223.3.txt, 223.patch1.txt, hive.223.4.patch, > patch.txt > > > today even when we do map side aggregates - we do multiple map-reduce jobs. > however - the reason for doing multiple map-reduce group-bys (for single > group-bys) was the fear of skews. When we are doing map side aggregates - > skews should not exist for the most part. There can be two reason for skews: > - large number of entries for a single grouping set - map side aggregates > should take care of this > - badness in hash function that sends too much stuff to one reducer - we > should be able to take care of this by having good hash functions (and prime > number reducer counts) > So i think we should be able to do a single stage map-reduce when doing > map-side aggregates. -- This message is automatically generated by JIRA. - You can reply to this email to add a comment to the issue online.
[jira] Updated: (HIVE-223) when using map-side aggregates - perform single map-reduce group-by
[ https://issues.apache.org/jira/browse/HIVE-223?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Namit Jain updated HIVE-223: Status: Patch Available (was: Open) added test for disabling map aggregation at run time > when using map-side aggregates - perform single map-reduce group-by > --- > > Key: HIVE-223 > URL: https://issues.apache.org/jira/browse/HIVE-223 > Project: Hadoop Hive > Issue Type: Improvement > Components: Query Processor >Reporter: Joydeep Sen Sarma >Assignee: Namit Jain > Attachments: 223.2.txt, 223.3.txt, 223.patch1.txt, hive.223.4.patch, > patch.txt > > > today even when we do map side aggregates - we do multiple map-reduce jobs. > however - the reason for doing multiple map-reduce group-bys (for single > group-bys) was the fear of skews. When we are doing map side aggregates - > skews should not exist for the most part. There can be two reason for skews: > - large number of entries for a single grouping set - map side aggregates > should take care of this > - badness in hash function that sends too much stuff to one reducer - we > should be able to take care of this by having good hash functions (and prime > number reducer counts) > So i think we should be able to do a single stage map-reduce when doing > map-side aggregates. -- This message is automatically generated by JIRA. - You can reply to this email to add a comment to the issue online.
[jira] Updated: (HIVE-223) when using map-side aggregates - perform single map-reduce group-by
[ https://issues.apache.org/jira/browse/HIVE-223?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Namit Jain updated HIVE-223: Attachment: patch.txt > when using map-side aggregates - perform single map-reduce group-by > --- > > Key: HIVE-223 > URL: https://issues.apache.org/jira/browse/HIVE-223 > Project: Hadoop Hive > Issue Type: Improvement > Components: Query Processor >Reporter: Joydeep Sen Sarma >Assignee: Namit Jain > Attachments: 223.2.txt, 223.3.txt, 223.patch1.txt, patch.txt > > > today even when we do map side aggregates - we do multiple map-reduce jobs. > however - the reason for doing multiple map-reduce group-bys (for single > group-bys) was the fear of skews. When we are doing map side aggregates - > skews should not exist for the most part. There can be two reason for skews: > - large number of entries for a single grouping set - map side aggregates > should take care of this > - badness in hash function that sends too much stuff to one reducer - we > should be able to take care of this by having good hash functions (and prime > number reducer counts) > So i think we should be able to do a single stage map-reduce when doing > map-side aggregates. -- This message is automatically generated by JIRA. - You can reply to this email to add a comment to the issue online.
[jira] Updated: (HIVE-223) when using map-side aggregates - perform single map-reduce group-by
[ https://issues.apache.org/jira/browse/HIVE-223?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Namit Jain updated HIVE-223: Status: Open (was: Patch Available) > when using map-side aggregates - perform single map-reduce group-by > --- > > Key: HIVE-223 > URL: https://issues.apache.org/jira/browse/HIVE-223 > Project: Hadoop Hive > Issue Type: Improvement > Components: Query Processor >Reporter: Joydeep Sen Sarma >Assignee: Namit Jain > Attachments: 223.2.txt, 223.3.txt, 223.patch1.txt, patch.txt > > > today even when we do map side aggregates - we do multiple map-reduce jobs. > however - the reason for doing multiple map-reduce group-bys (for single > group-bys) was the fear of skews. When we are doing map side aggregates - > skews should not exist for the most part. There can be two reason for skews: > - large number of entries for a single grouping set - map side aggregates > should take care of this > - badness in hash function that sends too much stuff to one reducer - we > should be able to take care of this by having good hash functions (and prime > number reducer counts) > So i think we should be able to do a single stage map-reduce when doing > map-side aggregates. -- This message is automatically generated by JIRA. - You can reply to this email to add a comment to the issue online.
[jira] Updated: (HIVE-223) when using map-side aggregates - perform single map-reduce group-by
[ https://issues.apache.org/jira/browse/HIVE-223?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Namit Jain updated HIVE-223: Status: Open (was: Patch Available) > when using map-side aggregates - perform single map-reduce group-by > --- > > Key: HIVE-223 > URL: https://issues.apache.org/jira/browse/HIVE-223 > Project: Hadoop Hive > Issue Type: Improvement > Components: Query Processor >Reporter: Joydeep Sen Sarma >Assignee: Namit Jain > Attachments: 223.2.txt, 223.3.txt, 223.patch1.txt > > > today even when we do map side aggregates - we do multiple map-reduce jobs. > however - the reason for doing multiple map-reduce group-bys (for single > group-bys) was the fear of skews. When we are doing map side aggregates - > skews should not exist for the most part. There can be two reason for skews: > - large number of entries for a single grouping set - map side aggregates > should take care of this > - badness in hash function that sends too much stuff to one reducer - we > should be able to take care of this by having good hash functions (and prime > number reducer counts) > So i think we should be able to do a single stage map-reduce when doing > map-side aggregates. -- This message is automatically generated by JIRA. - You can reply to this email to add a comment to the issue online.
[jira] Updated: (HIVE-223) when using map-side aggregates - perform single map-reduce group-by
[ https://issues.apache.org/jira/browse/HIVE-223?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Namit Jain updated HIVE-223: Attachment: 223.3.txt > when using map-side aggregates - perform single map-reduce group-by > --- > > Key: HIVE-223 > URL: https://issues.apache.org/jira/browse/HIVE-223 > Project: Hadoop Hive > Issue Type: Improvement > Components: Query Processor >Reporter: Joydeep Sen Sarma >Assignee: Namit Jain > Attachments: 223.2.txt, 223.3.txt, 223.patch1.txt > > > today even when we do map side aggregates - we do multiple map-reduce jobs. > however - the reason for doing multiple map-reduce group-bys (for single > group-bys) was the fear of skews. When we are doing map side aggregates - > skews should not exist for the most part. There can be two reason for skews: > - large number of entries for a single grouping set - map side aggregates > should take care of this > - badness in hash function that sends too much stuff to one reducer - we > should be able to take care of this by having good hash functions (and prime > number reducer counts) > So i think we should be able to do a single stage map-reduce when doing > map-side aggregates. -- This message is automatically generated by JIRA. - You can reply to this email to add a comment to the issue online.
[jira] Updated: (HIVE-223) when using map-side aggregates - perform single map-reduce group-by
[ https://issues.apache.org/jira/browse/HIVE-223?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Namit Jain updated HIVE-223: Status: Patch Available (was: Open) > when using map-side aggregates - perform single map-reduce group-by > --- > > Key: HIVE-223 > URL: https://issues.apache.org/jira/browse/HIVE-223 > Project: Hadoop Hive > Issue Type: Improvement > Components: Query Processor >Reporter: Joydeep Sen Sarma >Assignee: Namit Jain > Attachments: 223.2.txt, 223.3.txt, 223.patch1.txt > > > today even when we do map side aggregates - we do multiple map-reduce jobs. > however - the reason for doing multiple map-reduce group-bys (for single > group-bys) was the fear of skews. When we are doing map side aggregates - > skews should not exist for the most part. There can be two reason for skews: > - large number of entries for a single grouping set - map side aggregates > should take care of this > - badness in hash function that sends too much stuff to one reducer - we > should be able to take care of this by having good hash functions (and prime > number reducer counts) > So i think we should be able to do a single stage map-reduce when doing > map-side aggregates. -- This message is automatically generated by JIRA. - You can reply to this email to add a comment to the issue online.
[jira] Updated: (HIVE-223) when using map-side aggregates - perform single map-reduce group-by
[ https://issues.apache.org/jira/browse/HIVE-223?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Namit Jain updated HIVE-223: Status: Patch Available (was: Open) fixed a small bug > when using map-side aggregates - perform single map-reduce group-by > --- > > Key: HIVE-223 > URL: https://issues.apache.org/jira/browse/HIVE-223 > Project: Hadoop Hive > Issue Type: Improvement > Components: Query Processor >Reporter: Joydeep Sen Sarma >Assignee: Namit Jain > Attachments: 223.2.txt, 223.patch1.txt > > > today even when we do map side aggregates - we do multiple map-reduce jobs. > however - the reason for doing multiple map-reduce group-bys (for single > group-bys) was the fear of skews. When we are doing map side aggregates - > skews should not exist for the most part. There can be two reason for skews: > - large number of entries for a single grouping set - map side aggregates > should take care of this > - badness in hash function that sends too much stuff to one reducer - we > should be able to take care of this by having good hash functions (and prime > number reducer counts) > So i think we should be able to do a single stage map-reduce when doing > map-side aggregates. -- This message is automatically generated by JIRA. - You can reply to this email to add a comment to the issue online.
[jira] Updated: (HIVE-223) when using map-side aggregates - perform single map-reduce group-by
[ https://issues.apache.org/jira/browse/HIVE-223?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Namit Jain updated HIVE-223: Status: Open (was: Patch Available) > when using map-side aggregates - perform single map-reduce group-by > --- > > Key: HIVE-223 > URL: https://issues.apache.org/jira/browse/HIVE-223 > Project: Hadoop Hive > Issue Type: Improvement > Components: Query Processor >Reporter: Joydeep Sen Sarma >Assignee: Namit Jain > Attachments: 223.2.txt, 223.patch1.txt > > > today even when we do map side aggregates - we do multiple map-reduce jobs. > however - the reason for doing multiple map-reduce group-bys (for single > group-bys) was the fear of skews. When we are doing map side aggregates - > skews should not exist for the most part. There can be two reason for skews: > - large number of entries for a single grouping set - map side aggregates > should take care of this > - badness in hash function that sends too much stuff to one reducer - we > should be able to take care of this by having good hash functions (and prime > number reducer counts) > So i think we should be able to do a single stage map-reduce when doing > map-side aggregates. -- This message is automatically generated by JIRA. - You can reply to this email to add a comment to the issue online.
[jira] Updated: (HIVE-223) when using map-side aggregates - perform single map-reduce group-by
[ https://issues.apache.org/jira/browse/HIVE-223?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Namit Jain updated HIVE-223: Attachment: 223.2.txt > when using map-side aggregates - perform single map-reduce group-by > --- > > Key: HIVE-223 > URL: https://issues.apache.org/jira/browse/HIVE-223 > Project: Hadoop Hive > Issue Type: Improvement > Components: Query Processor >Reporter: Joydeep Sen Sarma >Assignee: Namit Jain > Attachments: 223.2.txt, 223.patch1.txt > > > today even when we do map side aggregates - we do multiple map-reduce jobs. > however - the reason for doing multiple map-reduce group-bys (for single > group-bys) was the fear of skews. When we are doing map side aggregates - > skews should not exist for the most part. There can be two reason for skews: > - large number of entries for a single grouping set - map side aggregates > should take care of this > - badness in hash function that sends too much stuff to one reducer - we > should be able to take care of this by having good hash functions (and prime > number reducer counts) > So i think we should be able to do a single stage map-reduce when doing > map-side aggregates. -- This message is automatically generated by JIRA. - You can reply to this email to add a comment to the issue online.
[jira] Updated: (HIVE-223) when using map-side aggregates - perform single map-reduce group-by
[ https://issues.apache.org/jira/browse/HIVE-223?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Namit Jain updated HIVE-223: Status: Patch Available (was: Open) fixed the bug that joy spotted, and also performed some code cleanup > when using map-side aggregates - perform single map-reduce group-by > --- > > Key: HIVE-223 > URL: https://issues.apache.org/jira/browse/HIVE-223 > Project: Hadoop Hive > Issue Type: Improvement > Components: Query Processor >Reporter: Joydeep Sen Sarma >Assignee: Namit Jain > Attachments: 223.2.txt, 223.patch1.txt > > > today even when we do map side aggregates - we do multiple map-reduce jobs. > however - the reason for doing multiple map-reduce group-bys (for single > group-bys) was the fear of skews. When we are doing map side aggregates - > skews should not exist for the most part. There can be two reason for skews: > - large number of entries for a single grouping set - map side aggregates > should take care of this > - badness in hash function that sends too much stuff to one reducer - we > should be able to take care of this by having good hash functions (and prime > number reducer counts) > So i think we should be able to do a single stage map-reduce when doing > map-side aggregates. -- This message is automatically generated by JIRA. - You can reply to this email to add a comment to the issue online.
[jira] Updated: (HIVE-223) when using map-side aggregates - perform single map-reduce group-by
[ https://issues.apache.org/jira/browse/HIVE-223?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Namit Jain updated HIVE-223: Status: Open (was: Patch Available) > when using map-side aggregates - perform single map-reduce group-by > --- > > Key: HIVE-223 > URL: https://issues.apache.org/jira/browse/HIVE-223 > Project: Hadoop Hive > Issue Type: Improvement > Components: Query Processor >Reporter: Joydeep Sen Sarma >Assignee: Namit Jain > Attachments: 223.patch1.txt > > > today even when we do map side aggregates - we do multiple map-reduce jobs. > however - the reason for doing multiple map-reduce group-bys (for single > group-bys) was the fear of skews. When we are doing map side aggregates - > skews should not exist for the most part. There can be two reason for skews: > - large number of entries for a single grouping set - map side aggregates > should take care of this > - badness in hash function that sends too much stuff to one reducer - we > should be able to take care of this by having good hash functions (and prime > number reducer counts) > So i think we should be able to do a single stage map-reduce when doing > map-side aggregates. -- This message is automatically generated by JIRA. - You can reply to this email to add a comment to the issue online.
[jira] Updated: (HIVE-223) when using map-side aggregates - perform single map-reduce group-by
[ https://issues.apache.org/jira/browse/HIVE-223?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Namit Jain updated HIVE-223: Status: Patch Available (was: Open) In a follow-up, I will do the following: In case of map-side aggregates, keep a count of hash table entries and the total entries on the map side. If the ratio is bad, assume that the user made a mistake, and switch to non map-side aggregation plan. > when using map-side aggregates - perform single map-reduce group-by > --- > > Key: HIVE-223 > URL: https://issues.apache.org/jira/browse/HIVE-223 > Project: Hadoop Hive > Issue Type: Improvement > Components: Query Processor >Reporter: Joydeep Sen Sarma >Assignee: Namit Jain > Attachments: 223.patch1.txt > > > today even when we do map side aggregates - we do multiple map-reduce jobs. > however - the reason for doing multiple map-reduce group-bys (for single > group-bys) was the fear of skews. When we are doing map side aggregates - > skews should not exist for the most part. There can be two reason for skews: > - large number of entries for a single grouping set - map side aggregates > should take care of this > - badness in hash function that sends too much stuff to one reducer - we > should be able to take care of this by having good hash functions (and prime > number reducer counts) > So i think we should be able to do a single stage map-reduce when doing > map-side aggregates. -- This message is automatically generated by JIRA. - You can reply to this email to add a comment to the issue online.
[jira] Updated: (HIVE-223) when using map-side aggregates - perform single map-reduce group-by
[ https://issues.apache.org/jira/browse/HIVE-223?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Namit Jain updated HIVE-223: Attachment: 223.patch1.txt > when using map-side aggregates - perform single map-reduce group-by > --- > > Key: HIVE-223 > URL: https://issues.apache.org/jira/browse/HIVE-223 > Project: Hadoop Hive > Issue Type: Improvement > Components: Query Processor >Reporter: Joydeep Sen Sarma >Assignee: Namit Jain > Attachments: 223.patch1.txt > > > today even when we do map side aggregates - we do multiple map-reduce jobs. > however - the reason for doing multiple map-reduce group-bys (for single > group-bys) was the fear of skews. When we are doing map side aggregates - > skews should not exist for the most part. There can be two reason for skews: > - large number of entries for a single grouping set - map side aggregates > should take care of this > - badness in hash function that sends too much stuff to one reducer - we > should be able to take care of this by having good hash functions (and prime > number reducer counts) > So i think we should be able to do a single stage map-reduce when doing > map-side aggregates. -- This message is automatically generated by JIRA. - You can reply to this email to add a comment to the issue online.
