On Monday, April 28, 2014 9:51:34 PM UTC+2, Otis Gospodnetic wrote:
>
> In short, yes. How much exactly depends on how intensive your indexing is.
>
Our indexing is pretty intensive since we have a number of machines sending
data as fast as they can to the Elasticsearch cluster.
So now we need to think about how to mitigate this. One idea is to add a
couple of machines to the cluster and then, assuming it is possible,
control the shard allocation so that the new index we are building gets
assigned to those new nodes and none of the other indexes do. That way the
new index is isolated to a couple of machines. We would of course direct
the indexing machines to talk directly to those machines.
Would this mitigate the performance impact on the rest of the cluster?
Does such a shard allocator exist? If not, is it possible to write one as a
plugin?
/MaF
--
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/ae101070-e2c3-4611-9a36-f380fd05534a%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.
