cassandra-commits mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Paul Loy (Commented) (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (CASSANDRA-2820) Re-introduce FastByteArrayInputStream (and Output equivalent)
Date Tue, 27 Sep 2011 15:06:11 GMT

    [ https://issues.apache.org/jira/browse/CASSANDRA-2820?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13115555#comment-13115555
] 

Paul Loy commented on CASSANDRA-2820:
-------------------------------------

yeah I think it would.

I've been looking at thrift for some things lately and that also uses the synchronized JDK
ByteArray*Streams and could possibly benefit from this too. Actually they do have a TByteArrayOutputStream
that allows them to get the byte[] buffer without a deep copy which may be of use to cassandra
also. So I could roll the two into one and contrib to commons-io. 
                
> Re-introduce FastByteArrayInputStream (and Output equivalent)
> -------------------------------------------------------------
>
>                 Key: CASSANDRA-2820
>                 URL: https://issues.apache.org/jira/browse/CASSANDRA-2820
>             Project: Cassandra
>          Issue Type: Improvement
>          Components: Core
>    Affects Versions: 0.8.0
>         Environment: n/a
>            Reporter: Paul Loy
>            Assignee: Paul Loy
>            Priority: Minor
>              Labels: bytearrayinputstream, bytearrayoutputstream, license, synchronized
>             Fix For: 1.0.0
>
>         Attachments: fast_bytearray_iostreams_harmony-patch-2.txt, fast_bytearray_iostreams_harmony-patch-3.txt,
fast_bytearray_iostreams_harmony-patch-4.txt, fast_bytearray_iostreams_harmony-patch-5.txt,
fast_bytearray_iostreams_harmony-patch-6.txt
>
>
> In https://issues.apache.org/jira/browse/CASSANDRA-37 FastByteArrayInputStream and FastByteArrayOutputStream
were removed due to being code copied from the JDK and then subsequently modified. The JDK
license is incompatible with Apache 2 license so the code had to go.
> I have since had a look at the performance of the JDK ByteArrayInputStream and a FastByteArrayInputStream
(i.e. one with synchronized methods made un-synchronized) and seen the difference is significant.
> After a warmup-period of >10000 loops I get the following for 10000 loops through
a 128000 byte array:
> bais : 3513ms
> fbais: 72ms
> This varies depending on the OS, machine and Java version, but it's always in favour
of the FastByteArrayInputStream as you might expect.
> Then, at Jonathan Ellis' suggestion, I tried this using a modified Apache Harmony ByteArrayInputStream
- i.e. one whose license is compatible - and the results were the same. A significant boost.
> I will attach a patch with changes for the 0.8.0 tag.

--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira

        

Mime
View raw message