Yes in general the fetch can be improved using standalone clients. I am NOT 
saying that data nodes are a bottleneck as of now.Indexing is not impacting 
the search. 

The point I am raising is data locality. Data is spread over few shards 
across few machines.Need to perform processing on this data without 
explicit fetch outside ES.
I want to return sample of grouped entries after going over all matched 
entries within an index.

Thanks,
Ram

On Saturday, December 13, 2014 9:24:11 PM UTC+5:30, Arie wrote:
>
> Hi,,
>
> Consider a non-data master node, this can improve data handling and search 
> speed a lot as I understand.
>
>

-- 
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 [email protected].
To view this discussion on the web visit 
https://groups.google.com/d/msgid/elasticsearch/4af3e662-f9a7-4200-84bf-4cc6af5e4d7d%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Reply via email to