You are viewing a plain text version of this content. The canonical link for it is here.
Posted to user@ignite.apache.org by rishi007bansod <ri...@gmail.com> on 2016/12/27 14:29:57 UTC

Performance of multiple ignite JVMs within single node

In my case database size in case of Onheap mode is 55 GB on single server
node. So from performance point of view which is preferable 
1. To have single ignite instance with heap size -Xmx > 55 GB 
2. Start multiple ignite JVM instances(of heap size 10 GBs) within single
node

Also does performance(query throughput) increases linearly as I increase
ignite JVM instances(for 2nd case)? 




--
View this message in context: http://apache-ignite-users.70518.x6.nabble.com/Performance-of-multiple-ignite-JVMs-within-single-node-tp9753.html
Sent from the Apache Ignite Users mailing list archive at Nabble.com.

Re: Performance of multiple ignite JVMs within single node

Posted by vkulichenko <va...@gmail.com>.
Option 1 is not usually good, because large heaps are likely to cause long GC
pauses. You should either start several smaller nodes, or use off-heap
memory [1].

As for query performance, there is a chance that you can improve *latency*
by starting several nodes (generally, as many nodes as many cores you have).
This is especially effective if queries imply large scans (e.g. aggregation
of the whole table, or other analytical use cases) and if the number of
parallel queries is not too big. However, if your system is highly loaded,
there are a lot of concurrent small queries and you're more interested in
throughput, this is not going to make much difference.

[1] https://apacheignite.readme.io/docs/off-heap-memory

-Val



--
View this message in context: http://apache-ignite-users.70518.x6.nabble.com/Performance-of-multiple-ignite-JVMs-within-single-node-tp9753p9758.html
Sent from the Apache Ignite Users mailing list archive at Nabble.com.

Re: Performance of multiple ignite JVMs within single node

Posted by "javastuff.sam@gmail.com" <ja...@gmail.com>.
I have observed significant degradation for get and put with increase in
number of Ignite JVM. I am still searching to tune this. 
 
More details can be found at -

http://apache-ignite-users.70518.x6.nabble.com/Performance-with-increase-in-node-td9378.html
<http://apache-ignite-users.70518.x6.nabble.com/Performance-with-increase-in-node-td9378.html>  

1 Server to 4 Server timing degraded ~7 fold for Put and 22 fold for fetch.
I see similar behavior with our application and with a simple test program.

-Sam




--
View this message in context: http://apache-ignite-users.70518.x6.nabble.com/Performance-of-multiple-ignite-JVMs-within-single-node-tp9753p9910.html
Sent from the Apache Ignite Users mailing list archive at Nabble.com.