Processors configuration (available_processors)

Elastic Stack Elasticsearch
1

I have a machine with 132GB of memory and 32 cores on which I am running
two elasticsearch nodes. Each node should have only half the total number
of CPU cores available so that both nodes can work at full capacity and not
block each other.
I believe the correct configuration option would be:

processors: 16

And I thought that this should change the reported value for

curl -XGET "http://localhost:9200/_nodes/os?pretty"

should be reporting the new value here:
"nodes" : {
"2GK7gBPNSSqqbRnRN_WmVg" : {
...
"os" : {
...
"available_processors" : 32, <-- I expect to see 16 here
...

Any ideas what I am doing wrong here, and how to set / confirm the number
of processors that an elasticsearch node should use.

Cheers,
-Robin-

--
You received this message because you are subscribed to the Google Groups "elasticsearch" group.
To unsubscribe from this group and stop receiving emails from it, send an email to elasticsearch+unsubscribe@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/elasticsearch/90cfb708-05db-4053-a289-ff0fbc561bbd%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

2

And in the spirit of answering your own questions, I've already been helped
to the answer:

curl -XGET "http://localhost:9200/_nodes/thread_pool?pretty
http://localhost:9200/_nodes/os?pretty"

...
"index" : {
"type" : "fixed",
"min" : 16,
"max" : 16,
"queue_size" : "200"
...

It seems this cannot be set using the config API

Cheers,
-Robin-

On Tuesday, 16 December 2014 12:04:03 UTC+1, Robin Clarke wrote:

I have a machine with 132GB of memory and 32 cores on which I am running
two elasticsearch nodes. Each node should have only half the total number
of CPU cores available so that both nodes can work at full capacity and not
block each other.
I believe the correct configuration option would be:

processors: 16

And I thought that this should change the reported value for

curl -XGET "http://localhost:9200/_nodes/os?pretty"

should be reporting the new value here:
"nodes" : {
"2GK7gBPNSSqqbRnRN_WmVg" : {
...
"os" : {
...
"available_processors" : 32, <-- I expect to see 16 here
...

Any ideas what I am doing wrong here, and how to set / confirm the number
of processors that an elasticsearch node should use.

Cheers,
-Robin-

--
You received this message because you are subscribed to the Google Groups "elasticsearch" group.
To unsubscribe from this group and stop receiving emails from it, send an email to elasticsearch+unsubscribe@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/elasticsearch/abd0aa9f-2f2e-45b0-8822-00c64160c34d%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

3

You have to set up a container and assign virtual processors to it, and
install ES in there so the Java JVM can use only these cores.

With 16 cores, you have 32 processors because of Intel's hyperthreading
technology which doubles the number of available processing units in a core.

Thread pool is a different topic.

Jörg

On Tue, Dec 16, 2014 at 12:13 PM, Robin Clarke robin13@gmail.com wrote:

And in the spirit of answering your own questions, I've already been
helped to the answer:

curl -XGET "http://localhost:9200/_nodes/thread_pool?pretty
http://localhost:9200/_nodes/os?pretty"

...
"index" : {
"type" : "fixed",
"min" : 16,
"max" : 16,
"queue_size" : "200"
...

It seems this cannot be set using the config API

Cheers,
-Robin-

On Tuesday, 16 December 2014 12:04:03 UTC+1, Robin Clarke wrote:

I have a machine with 132GB of memory and 32 cores on which I am running
two elasticsearch nodes. Each node should have only half the total number
of CPU cores available so that both nodes can work at full capacity and not
block each other.
I believe the correct configuration option would be:

processors: 16

And I thought that this should change the reported value for

curl -XGET "http://localhost:9200/_nodes/os?pretty"

should be reporting the new value here:
"nodes" : {
"2GK7gBPNSSqqbRnRN_WmVg" : {
...
"os" : {
...
"available_processors" : 32, <-- I expect to see 16 here
...

Any ideas what I am doing wrong here, and how to set / confirm the number
of processors that an elasticsearch node should use.

Cheers,
-Robin-

--
You received this message because you are subscribed to the Google Groups
"elasticsearch" group.
To unsubscribe from this group and stop receiving emails from it, send an
email to elasticsearch+unsubscribe@googlegroups.com.
To view this discussion on the web visit
https://groups.google.com/d/msgid/elasticsearch/abd0aa9f-2f2e-45b0-8822-00c64160c34d%40googlegroups.com
https://groups.google.com/d/msgid/elasticsearch/abd0aa9f-2f2e-45b0-8822-00c64160c34d%40googlegroups.com?utm_medium=email&utm_source=footer
.

For more options, visit https://groups.google.com/d/optout.

--
You received this message because you are subscribed to the Google Groups "elasticsearch" group.
To unsubscribe from this group and stop receiving emails from it, send an email to elasticsearch+unsubscribe@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/elasticsearch/CAKdsXoH9SKwrek6Zv7KHQPRvziF_jeQFK-533L23pgErSidepA%40mail.gmail.com.
For more options, visit https://groups.google.com/d/optout.