[
https://issues.apache.org/jira/browse/HADOOP-10674?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14042769#comment-14042769
]
Jing Zhao commented on HADOOP-10674:
------------------------------------
It will be better to change the following code
{code}
+ for(final int end = offset + len - remainder; i < end; i += 8) {
+ final int x = localCrc
+ ^ (((b[i ] << 24) >>> 24) + ((b[i+1] << 24) >>> 16))
+ + (((b[i+2] << 24) >>> 8 ) + (b[i+3] << 24));
{code}
to
{code}
+ for(final int end = offset + len - remainder; i < end; i += 8) {
+ final int x = localCrc
+ ^ (((b[i ] << 24) >>> 24) + ((b[i+1] << 24) >>> 16))
+ + ((b[i+2] << 24) >>> 8 ) + (b[i+3] << 24)));
{code}
Other than that +1.
> Rewrite the PureJavaCrc32 loop for performance improvement
> ----------------------------------------------------------
>
> Key: HADOOP-10674
> URL: https://issues.apache.org/jira/browse/HADOOP-10674
> Project: Hadoop Common
> Issue Type: Improvement
> Components: performance, util
> Reporter: Tsz Wo Nicholas Sze
> Assignee: Tsz Wo Nicholas Sze
> Attachments: c10674_20140609.patch, c10674_20140609b.patch,
> c10674_20140610.patch, c10674_20140612.patch, c10674_20140619.patch
>
>
> Below are some performance improvement opportunities performance improvement
> in PureJavaCrc32.
> - eliminate "off += 8; len -= 8;"
> - replace T8_x_start with hard coded constants
> - eliminate c0 - c7 local variables
> In my machine, there are 30% to 50% improvement for most of the cases.
--
This message was sent by Atlassian JIRA
(v6.2#6252)
