hadoop-pig-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Daniel Dai (JIRA)" <j...@apache.org>
Subject [jira] Updated: (PIG-1336) Optimize POStore serialized into JobConf
Date Tue, 30 Mar 2010 06:22:27 GMT

     [ https://issues.apache.org/jira/browse/PIG-1336?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]

Daniel Dai updated PIG-1336:
----------------------------

    Description: We serialize POStore too early in the JobControlCompiler. At that time, storeFunc
have unconstraint link to other operator; in the worst case, it will chain the whole physical
plan. Also, in multi-store case, POStore has link to its data source, which is not needed
and will increase the footprint of serialized POStore.   (was: Currently, if a pig script
job contains multiple map-reduce jobs, each job will serialize all map-reduce job plan into
JobConf. The reason is PhysicalOperator.inputs is built in Physical plan and in fact it chains
all the jobs without regard to map-reduce boundary. Further, when we only want to serialize
POStore, we serialize this whole plan again due to POStore.inputs. This should be fixed. )
        Summary: Optimize POStore serialized into JobConf  (was: Optimize content of mapPlan/reducePlan
to be serialized into JobConf)

> Optimize POStore serialized into JobConf
> ----------------------------------------
>
>                 Key: PIG-1336
>                 URL: https://issues.apache.org/jira/browse/PIG-1336
>             Project: Pig
>          Issue Type: Improvement
>          Components: impl
>    Affects Versions: 0.7.0
>            Reporter: Daniel Dai
>            Assignee: Daniel Dai
>         Attachments: PIG-1336-1.patch, PIG-1336-2.patch
>
>
> We serialize POStore too early in the JobControlCompiler. At that time, storeFunc have
unconstraint link to other operator; in the worst case, it will chain the whole physical plan.
Also, in multi-store case, POStore has link to its data source, which is not needed and will
increase the footprint of serialized POStore. 

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.


Mime
View raw message