[
https://issues.apache.org/jira/browse/HBASE-9956?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13821741#comment-13821741
]
Hudson commented on HBASE-9956:
-------------------------------
SUCCESS: Integrated in hbase-0.96 #188 (See
[https://builds.apache.org/job/hbase-0.96/188/])
HBASE-9956 Remove keyLength cache from KeyValue (larsh: rev 1541313)
*
/hbase/branches/0.96/hbase-common/src/main/java/org/apache/hadoop/hbase/KeyValue.java
*
/hbase/branches/0.96/hbase-server/src/main/java/org/apache/hadoop/hbase/io/hfile/HFileReaderV2.java
*
/hbase/branches/0.96/hbase-server/src/main/java/org/apache/hadoop/hbase/regionserver/ScanQueryMatcher.java
> Remove keyLength cache from KeyValue
> ------------------------------------
>
> Key: HBASE-9956
> URL: https://issues.apache.org/jira/browse/HBASE-9956
> Project: HBase
> Issue Type: Bug
> Reporter: Lars Hofhansl
> Assignee: Lars Hofhansl
> Priority: Minor
> Fix For: 0.98.0, 0.96.1, 0.94.14
>
> Attachments: 9935-0.94.txt, 9956-0.94.txt, 9956-trunk.txt
>
>
> See discussion in HBASE-9935. Currently KeyValue caches the keyLength in
> order to avoid decoding the key length in getKeyLength() repeatedly.
> Benchmarking outside of a profiler revealed no improvement from that cache
> (and a sampling profiler also does not indicate getKeyLength() as a
> performance issue).
> Along with this there's also a slight fix in ScanQueryMatcher.
--
This message was sent by Atlassian JIRA
(v6.1#6144)
