[
https://issues.apache.org/jira/browse/HBASE-17877?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15967397#comment-15967397
]
Vikas Vishwakarma edited comment on HBASE-17877 at 4/13/17 12:03 PM:
---------------------------------------------------------------------
[~anoop.hbase] will take a look into BBUtils also.
[~larsh] one quick doubt. The HBase byte array comparator gives only 3
possible output for any byte array input -1, 0, +1
With Guava and Hadoop comparator the output can be any -integer, 0 , +integer
for example , for this input
Arrays.fill( ba1_8, (byte) 43 );
Arrays.fill( ba2_8, (byte) 43 );
ba1_8[0] = (byte) 32;
ba2_8[0] = (byte) 43;
HBase comparator will give output as -1 while Guava & Hadoop comparator will
give output as -11
Mostly functions should check comparator output as > < or == , this should be a
fair assumption or do we need to map the guava comparator output in the -1, 0,
1 range?
was (Author: vik.karma):
[~anoop.hbase] will take a look into BBUtils also.
[~larsh] one quick doubt. The HBase byte array comparator gives only 3
possible output for any byte array input -1, 0, +1
With Guava and Hadoop comparator the output can be any -integer, 0 , +integer
for example , for this input
Arrays.fill( ba1_8, (byte) 43 );
Arrays.fill( ba2_8, (byte) 43 );
ba1_8[0] = (byte) 32;
ba2_8[0] = (byte) 43;
HBase comparator will give output as -1 while Guava & Hadoop comparator will
give output as -11
Mostly functions should check comparator output as > < or == , this should be a
fair assumption or do we need to do a thorough verification around this?
> Replace/improve HBase's byte[] comparator
> -----------------------------------------
>
> Key: HBASE-17877
> URL: https://issues.apache.org/jira/browse/HBASE-17877
> Project: HBase
> Issue Type: Bug
> Reporter: Lars Hofhansl
> Assignee: Vikas Vishwakarma
> Attachments: 17877-1.2.patch, 17877-v2-1.3.patch,
> ByteComparatorJiraHBASE-17877.pdf
>
>
> [~vik.karma] did some extensive tests and found that Hadoop's version is
> faster - dramatically faster in some cases.
> Patch forthcoming.
--
This message was sent by Atlassian JIRA
(v6.3.15#6346)
