hbase-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "stack (JIRA)" <j...@apache.org>
Subject [jira] Updated: (HBASE-553) Remove bloom filters from HBase when they become available in Hadoop
Date Mon, 15 Dec 2008 22:31:44 GMT

     [ https://issues.apache.org/jira/browse/HBASE-553?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]

stack updated HBASE-553:
------------------------

    Fix Version/s: 0.20.0

Marking as fix for 0.20.0.

Andzrej's patch, HADOOP-3063, has been committed to hadoop TRUNK.  Will be available in hadoop
0.20.0.   Of note, Andrzej uses the DyamicBloomFilter which resizes as inserts rise.  We should
do same if we push on with current implementation though, I think it should be possible for
us to know at time of writing how many keys will make it into a flush file.  We won't know
how many instances on a particular row or particular row/column combinations in advance so
maybe a DBF makes sense anyways.

> Remove bloom filters from HBase when they become available in Hadoop
> --------------------------------------------------------------------
>
>                 Key: HBASE-553
>                 URL: https://issues.apache.org/jira/browse/HBASE-553
>             Project: Hadoop HBase
>          Issue Type: Improvement
>          Components: util
>    Affects Versions: 0.2.0
>            Reporter: Jim Kellerman
>            Priority: Minor
>             Fix For: 0.20.0
>
>
> Hadoop is incorporating the bloom filter code that is in HBase to make a BloomFilterMapFile
(HADOOP-3063). Once this is committed to Hadoop trunk, the bloom filter code can be removed
from HBase.

-- 
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