[
https://issues.apache.org/jira/browse/IGNITE-6699?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16420420#comment-16420420
]
Ivan Fedotov commented on IGNITE-6699:
--------------------------------------
There are results of benchmarks on streaming 7000 integers with 1 client and 2
server nodes:
*addData(Collection)* -
https://gist.github.com/1vanan/8a1d310526ba3783894f5181067a272c
*addData(Key, Value)* master -
https://gist.github.com/1vanan/cbe10aed7768b00fa6e4176b7e6ed51c
*addData(Key, Value)* ignite-6699 -
https://gist.github.com/1vanan/cef0abf262bcc054aaa9ec54d84bb4f1
According to result, the time of data streaming per key/value has decreased on
*32%* in average and became equal to streaming data per collection. The smaller
data streaming time is better.
> Optimize client-side data streamer performance
> ----------------------------------------------
>
> Key: IGNITE-6699
> URL: https://issues.apache.org/jira/browse/IGNITE-6699
> Project: Ignite
> Issue Type: Task
> Components: streaming
> Affects Versions: 2.3
> Reporter: Vladimir Ozerov
> Assignee: Ivan Fedotov
> Priority: Major
> Labels: iep-1, performance
> Fix For: 2.5
>
>
> Currently if a user has several server nodes and a single client node with
> single thread pushing data to streamer, he will not be able to load data at
> maximum speed. On the other hand, if he start several data loading threads,
> throughput will increase.
> One of root causes of this is bad data streamer design. Method
> {{IgniteDataStreamer.addData(K, V)}} returns new feature for every operation,
> this is too fine grained approach. Also it generates a lot of garbage and
> causes contention on streamer internals.
> Proposed implementation flow:
> 1) Compare performance of {{addData(K, V)}} vs {{addData(Collection)}}
> methods from one thread in distributed environment. The latter should show
> considerably higher throughput.
> 2) Users should receive per-batch features, rather than per-key.
> 3) Try caching thread data in some collection until it is large enough to
> avoid contention and unnecessary allocations.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
