hive-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Wei Zheng (JIRA)" <j...@apache.org>
Subject [jira] [Updated] (HIVE-9277) Hybrid Hybrid Grace Hash Join
Date Sat, 21 Mar 2015 01:26:38 GMT

     [ https://issues.apache.org/jira/browse/HIVE-9277?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]

Wei Zheng updated HIVE-9277:
----------------------------
    Attachment: HIVE-9277.15.patch

Upload 15th patch for testing.

If this one passes tests, it is ready to be committed. Thanks.

> Hybrid Hybrid Grace Hash Join
> -----------------------------
>
>                 Key: HIVE-9277
>                 URL: https://issues.apache.org/jira/browse/HIVE-9277
>             Project: Hive
>          Issue Type: New Feature
>          Components: Physical Optimizer
>            Reporter: Wei Zheng
>            Assignee: Wei Zheng
>              Labels: join
>         Attachments: HIVE-9277.01.patch, HIVE-9277.02.patch, HIVE-9277.03.patch, HIVE-9277.04.patch,
HIVE-9277.05.patch, HIVE-9277.06.patch, HIVE-9277.07.patch, HIVE-9277.08.patch, HIVE-9277.13.patch,
HIVE-9277.14.patch, HIVE-9277.15.patch, High-leveldesignforHybridHybridGraceHashJoinv1.0.pdf
>
>
> We are proposing an enhanced hash join algorithm called _“hybrid hybrid grace hash
join”_.
> We can benefit from this feature as illustrated below:
> * The query will not fail even if the estimated memory requirement is slightly wrong
> * Expensive garbage collection overhead can be avoided when hash table grows
> * Join execution using a Map join operator even though the small table doesn't fit in
memory as spilling some data from the build and probe sides will still be cheaper than having
to shuffle the large fact table
> The design was based on Hadoop’s parallel processing capability and significant amount
of memory available.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

Mime
View raw message