ignite-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From ezhuravlev <e.zhuravlev...@gmail.com>
Subject Re: Task management - MapReduce & ForkJoin performance penalty
Date Fri, 01 Sep 2017 17:48:43 GMT

I've added Thread.sleep(200) to Jobs to simulate a small load.
Here is what I've got: 
1node: 1 Task 2000 Jobs ~25 sec
2nodes(on the same machine): 1 Task 2000 Jobs ~13 sec

What I want to say here - this overhead will be not noticeable on real Jobs.

What about some configuration changes - you could change
igniteConfiguration.setPublicThreadPoolSize(). By default, it set to CPU
count(with hyperthreading), but for small tasks you can make it bigger, for
example, CPU count*2.

Also, Ignite contains module benchmarks - you can check it and find
benchmarks for ComputeGrid.

All the best,

Sent from: http://apache-ignite-users.70518.x6.nabble.com/

View raw message