[
https://issues.apache.org/jira/browse/IGNITE-6699?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16306229#comment-16306229
]
Ivan Fedotov commented on IGNITE-6699:
--------------------------------------
[~vozerov] Hello!
I've wrote benchmarks on methods addData(Key, Value) & addData(Collection) and
got the next results:
Benchmark Mode Cnt Score Error Units
JmhStreamerAddDataBenchmark.*addDataCollection* avgt 20 *351,842* ± 41,470 us/op
JmhStreamerAddDataBenchmark.*addDataKeyValue* avgt 20 *525,112* ± 44,872 us/op
There are expected results: loading data with collection is more than 0.6 times
faster than loading data through key/value.
I'll try to catch Futures in GridCompoundFuture and load when it reaches
necessary size.
> 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
> Labels: iep-1, performance
> Fix For: 2.4
>
>
> 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
(v6.4.14#64029)
