[
https://issues.apache.org/jira/browse/HBASE-20526?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16464628#comment-16464628
]
Key Hutu commented on HBASE-20526:
----------------------------------
In the application, doBulkload(hpath, admin, table, regionLocator) method
called.
To ensure real-time performance, many small files at higher frequencies was
loaded
> multithreads bulkload performance
> ---------------------------------
>
> Key: HBASE-20526
> URL: https://issues.apache.org/jira/browse/HBASE-20526
> Project: HBase
> Issue Type: Improvement
> Components: mapreduce, Zookeeper
> Affects Versions: 1.2.0
> Environment: hbase-server-1.2.0-cdh5.12.1
> spark version 1.6
> Reporter: Key Hutu
> Assignee: Key Hutu
> Priority: Minor
> Labels: performance
> Original Estimate: 96h
> Remaining Estimate: 96h
>
> When doing bulkload , some interactive with zookeeper to getting region key
> range may be cost more time.
> In multithreads enviorment, the duration maybe cost 5 minute or more.
> From the executor log, like 'Reading reply sessionid:0x262fb37f4a07080 ,
> packet:: clientPath:null server ...' contents appear many times.
>
> It likely to provide new method for bulkload, caching the key range outside
>
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
