[ https://issues.apache.org/jira/browse/HBASE-4012?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13936041#comment-13936041 ]
Manukranth Kolloju commented on HBASE-4012: ------------------------------------------- Yes, In the above paste, prefix ~ Common prefix. The actual rows can be arbitrarily long. It is reasonable to have very good performance in case of long prefixes. But I was curious if someone else also did study the short prefixes case. Especially where the common prefix was comparable to length of long. I was groping to see parity in those cases but instead there was a regression. As far as my simple benchmarks show. > Further optimize byte comparison methods > ---------------------------------------- > > Key: HBASE-4012 > URL: https://issues.apache.org/jira/browse/HBASE-4012 > Project: HBase > Issue Type: Improvement > Components: util > Affects Versions: 0.92.0 > Reporter: Todd Lipcon > Assignee: Ted Yu > Priority: Minor > Labels: noob > Fix For: 0.92.0 > > Attachments: 4012-v2.txt, 4012.txt > > > Guava uses some clever tricks with sun.misc.Unsafe to compare byte arrays > about 100% faster than the naive byte-by-byte implementation: > http://guava-libraries.googlecode.com/svn/trunk/guava/src/com/google/common/primitives/UnsignedBytes.java > We should borrow this [Apache 2 licensed] code. -- This message was sent by Atlassian JIRA (v6.2#6252)