Machine learning toolkit performance impacts on the cluster

[nawazns5038]

Hi,

For installing the Machine Learning Toolkit on a search head cluster, are there any impacts on the performance of the cluster.
Is it necessary to install the addon on the indexer cluster as well.

[adonio]

hello there:

per docs:
http://docs.splunk.com/Documentation/MLApp/3.1.0/User/Installandconfigure#Distributed_deployment
you have to have Splunk 6.4 or greater for leveraging ML toolkit on SHC
regarding Indexer Cluster, same doc link.
"install Python for Scientific Computing on the indexers in your cluster. This feature is disabled by default."

hope it helps

[nawazns5038]

Hi,

Is it a must to install the addon on the indexer cluster .

Problem replicating config (bundle) to search peer '10.x.x.xx.xx:8089', error while transmitting bundle data.

I am getting that error on the search head after I enable the app on the search head without an addon on the indexer cluster. How can this be resolved and why is the bundle being pushed to the indexers while I install in the search head .

Thanks,
N

[adonio]

according to the link in the answer, "If you want to use the distributed apply feature of the Splunk Machine Learning Toolkit, install Python for Scientific Computing on the indexers in your cluster.
not sure what add-o regarding the error, or why bundles pushed to indexers while you install on SH.
if you have a SHC - install on Deployer and push to members

[nawazns5038]

I just found this link and you can go through it .

http://docs.splunk.com/Documentation/Splunk/6.3.1/DistSearch/Whatsearchheadssend

It does not say that it is mandatory to install on indexers . I'll try installing the addon on the indexers and check back for errors.

[adonio]

please pay attention to splunk versions and the docs versions...

[nawazns5038]

yes, our search head is 6.5.3 .. and the same doc all versions