apex-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "ASF GitHub Bot (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (APEXCORE-606) Suggestion: Optimise Kryo Output
Date Thu, 26 Jan 2017 19:38:24 GMT

    [ https://issues.apache.org/jira/browse/APEXCORE-606?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15840291#comment-15840291
] 

ASF GitHub Bot commented on APEXCORE-606:
-----------------------------------------

GitHub user brightchen opened a pull request:

    https://github.com/apache/apex-core/pull/455

    APEXCORE-606-RV: review only: optimise kryo output stream.

    This is a prototype of optimize kryo output stream. The main focus is avoiding memory
copy and garbage collection to increase the performance.
    - SerializationBuffer is the the class extends kryo Output which use the WindowedBlockStream
to manage the memory.
    - WindowedBlockStream is the memory management class which support manage memory blocks
based windows. The windowed related function can be ignored now
    - BlockStream is the super class of WindowedBlockStream which manage blocks
    - CodecPerformanceTest is the class which test the performance. Please change the CodecPerformanceTest.loop
to a number bigger than 100 before really do the performance test. And test the use case one
by one as initialize impact the performance.
    - PartitionSerde is the class serialize partition and object which can be used by BufferServerPublisher.
    - BufferServerPublisher.put_new(Object) is the implementation use new stream.

You can merge this pull request into a Git repository by running:

    $ git pull https://github.com/brightchen/apex-core APEXCORE-606-RV

Alternatively you can review and apply these changes as the patch at:

    https://github.com/apache/apex-core/pull/455.patch

To close this pull request, make a commit to your master/trunk branch
with (at least) the following in the commit message:

    This closes #455
    
----
commit a540faa7f60fc220149cf087efee0ac3698ff2ce
Author: brightchen <bright@datatorrent.com>
Date:   2016-12-27T23:09:11Z

    APEXCORE-606-RV: review only: optimise kryo output stream.

----


> Suggestion: Optimise Kryo Output
> --------------------------------
>
>                 Key: APEXCORE-606
>                 URL: https://issues.apache.org/jira/browse/APEXCORE-606
>             Project: Apache Apex Core
>          Issue Type: New Feature
>            Reporter: bright chen
>            Assignee: bright chen
>
> The kryo Output has some limitation
>   - The size of the data is limited. kryo write data to the buffer, it will throw the
overflow exception if the data exceed the size
>   - The Output.toBytes() will copy the data to temporary buffer and output, it will 
decrease the performance and introduce garbage collection.
> When I was tuning Spillable Data structure and Manage State. I create a mechanism to
share and reuse the memory to avoid above problem.  And it can be reused in core serialization
with small change. Please see jira: https://issues.apache.org/jira/browse/APEXMALHAR-2190



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

Mime
View raw message