ignite-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Andrey Gura <ag...@apache.org>
Subject Re: accurate offheap metrics
Date Wed, 08 Mar 2017 16:54:30 GMT

JVM allocates more memory than pointed out by Xms and Xmx parameters.
For my case is 2-4 Gb more allocated for 4-8 Gb heaps. I think that it
is related with preloaded classes for faster start new JVM processes.

Ignite internal structures that provide offheap cache also allocates
memory in off heap but MBean doesn't take it into account. There are
two structures: offheap map implementation and offheap LRU eviction
policy implementation. Both structures of course are created for each
off heap cache. There is issue about this problem [1]

[1] https://issues.apache.org/jira/browse/IGNITE-4797

On Tue, Mar 7, 2017 at 6:59 PM, lawrencefinn <larryf@gmail.com> wrote:
> What's the best way to get accurate metrics of offheap memory used?  I've
> tried a bunch of different ways such as utilizing jmx, utilizing the API and
> iterating through each cache on each node to get offHeapAllocatedSize, using
> rest api, etc...  None of the values add up right.  I see my ignite process
> taking up 4GB, max heap is 2GB, and all the calculations are around 300MB.
> Are indexes not taken into account?
> --
> View this message in context: http://apache-ignite-users.70518.x6.nabble.com/accurate-offheap-metrics-tp11057.html
> Sent from the Apache Ignite Users mailing list archive at Nabble.com.

View raw message