hbase-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From st...@duboce.net
Subject Re: Review Request: Batch writes should sync to HLog in batches
Date Thu, 17 Jun 2010 05:23:01 GMT


> On 2010-06-16 20:24:22, Kannan Muthukkaruppan wrote:
> > Todd: I know this is already reviewed and committed. I just got around to reviewing
this and the related HBase-2733. Looks really good!!
> > 
> > One side effect of this change is that now a single WALEdit could contain data for
more than 1 row. Previously all KVs inside a WALEdit corresponded to the same row. I don't
think it should cause any problems, but thought I should call it out in case others can think
of potential issues.
> > 
> > 
> >

I think its going to be alright Kannan.  The patch over in hbase-1025, the fixup to the replay
of split edits, is applying the kvs it finds in a WALEdit value one at a time.  That they
are for different rows but of the same region, it should be fine I'd say.


- stack


-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
http://review.hbase.org/r/167/#review247
-----------------------------------------------------------


On 2010-06-11 15:18:10, Todd Lipcon wrote:
> 
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> http://review.hbase.org/r/167/
> -----------------------------------------------------------
> 
> (Updated 2010-06-11 15:18:10)
> 
> 
> Review request for hbase, Kannan Muthukkaruppan and Ryan Rawson.
> 
> 
> Summary
> -------
> 
> I implemented the "mini batching" idea we talked about on the JIRA.
> 
> This currently breaks some of the error handling, so I dont intend to commit as is, but
everyone is busy so wanted to put a review up now while I tidy up the rest.
> 
> 
> This addresses bug HBASE-2353.
>     http://issues.apache.org/jira/browse/HBASE-2353
> 
> 
> Diffs
> -----
> 
>   src/main/java/org/apache/hadoop/hbase/HConstants.java 1e59533 
>   src/main/java/org/apache/hadoop/hbase/ipc/HRegionInterface.java 62617ac 
>   src/main/java/org/apache/hadoop/hbase/regionserver/HRegion.java adc505b 
>   src/main/java/org/apache/hadoop/hbase/regionserver/HRegionServer.java 541ec9b 
>   src/main/java/org/apache/hadoop/hbase/regionserver/wal/HLog.java ba04820 
>   src/test/java/org/apache/hadoop/hbase/MultithreadedTestUtil.java 870f925 
>   src/test/java/org/apache/hadoop/hbase/TestAcidGuarantees.java 8a5206c 
>   src/test/java/org/apache/hadoop/hbase/regionserver/TestHRegion.java e75767d 
> 
> Diff: http://review.hbase.org/r/167/diff
> 
> 
> Testing
> -------
> 
> Some PEs on a real sync-enabled cluster, seems faster but haven't done scientific benchmarking.
> 
> 
> Thanks,
> 
> Todd
> 
>


Mime
View raw message