We are running a node client on each machine with small JVM heap -Xms384m
-Xmx384m -Xss256k which is suitable for our use case.
There are however another 285MB non-heap memory (676-389=285).
How can this extra non-heap memory usage be configured ? What is it used
for ?
Below are the relevant node stats.
Locking heap memory usage is done by Elasticsearch on data nodes to reduce
GC rounds, mainly because it loads a lot of data that is best managed by ES
itself.
On client nodes you don't need that (and if you did, you wouldn't be using
that small heap sizes)
We are running a node client on each machine with small JVM heap -Xms384m
-Xmx384m -Xss256k which is suitable for our use case.
There are however another 285MB non-heap memory (676-389=285).
How can this extra non-heap memory usage be configured ? What is it used
for ?
Below are the relevant node stats.
I would like to reduce the amount of non-heap memory used by a client node.
I would like to reclaim as much as I can from these 280MB, what is it used
for?
On Tuesday, January 13, 2015 at 12:56:32 PM UTC+2, Itamar Syn-Hershko wrote:
Why would you want that?
Locking heap memory usage is done by Elasticsearch on data nodes to reduce
GC rounds, mainly because it loads a lot of data that is best managed by ES
itself.
On client nodes you don't need that (and if you did, you wouldn't be using
that small heap sizes)
On Tue, Jan 13, 2015 at 12:37 PM, Itai Frenkel <itaif...@live.com
<javascript:>> wrote:
Hello,
We are running a node client on each machine with small JVM heap -Xms384m
-Xmx384m -Xss256k which is suitable for our use case.
There are however another 285MB non-heap memory (676-389=285).
How can this extra non-heap memory usage be configured ? What is it used
for ?
Below are the relevant node stats.
I would like to reduce the amount of non-heap memory used by a client
node. I would like to reclaim as much as I can from these 280MB, what is it
used for?
On Tuesday, January 13, 2015 at 12:56:32 PM UTC+2, Itamar Syn-Hershko
wrote:
Why would you want that?
Locking heap memory usage is done by Elasticsearch on data nodes to
reduce GC rounds, mainly because it loads a lot of data that is best
managed by ES itself.
On client nodes you don't need that (and if you did, you wouldn't be
using that small heap sizes)
On Tue, Jan 13, 2015 at 12:37 PM, Itai Frenkel itaif...@live.com wrote:
Hello,
We are running a node client on each machine with small JVM
heap -Xms384m -Xmx384m -Xss256k which is suitable for our use case.
There are however another 285MB non-heap memory (676-389=285).
How can this extra non-heap memory usage be configured ? What is it used
for ?
Below are the relevant node stats.
Apache, Apache Lucene, Apache Hadoop, Hadoop, HDFS and the yellow elephant
logo are trademarks of the
Apache Software Foundation
in the United States and/or other countries.