I have a requirement to analyze on how elasticsearch be used in
federation.
Let me try to explain;
Lets say, as a part of a big enterprise, I have a ES cluster - ES001
with 3 nodes, using the default configurations of 1 replica and 2
shards deployed in LA, USA. Another ES cluster - ES002 with the same
configuration as ES001 deployed at Cork, Ireland.
Is there a way to synchronize between these 2 clusters? I'm not
talking about data replication, can a search query on cluster ES001 be
directed to ES002 based on some parameter?
?. You code issues the search request, you can decide which one to hit.
On Thursday, March 1, 2012 at 9:02 AM, ajan wrote:
I have a requirement to analyze on how elasticsearch be used in
federation.
Let me try to explain;
Lets say, as a part of a big enterprise, I have a ES cluster - ES001
with 3 nodes, using the default configurations of 1 replica and 2
shards deployed in LA, USA. Another ES cluster - ES002 with the same
configuration as ES001 deployed at Cork, Ireland.
Is there a way to synchronize between these 2 clusters? I'm not
talking about data replication, can a search query on cluster ES001 be
directed to ES002 based on some parameter?
I suppose you want to run queries as if it was one cluster, but insert
data only on the local nodes, and not migrate data between different
locations.
If that's the case, you could also search both clusters and aggregate
the results in your app. Or, you can put all nodes in the same ES
cluster, and use shard allocation to make sure local data gets indexed
in local nodes.
In case you find out other options, I'd be really interested
?. You code issues the search request, you can decide which one to hit.
On Thursday, March 1, 2012 at 9:02 AM, ajan wrote:
I have a requirement to analyze on how elasticsearch be used in
federation.
Let me try to explain;
Lets say, as a part of a big enterprise, I have a ES cluster - ES001
with 3 nodes, using the default configurations of 1 replica and 2
shards deployed in LA, USA. Another ES cluster - ES002 with the same
configuration as ES001 deployed at Cork, Ireland.
Is there a way to synchronize between these 2 clusters? I'm not
talking about data replication, can a search query on cluster ES001 be
directed to ES002 based on some parameter?
I suppose you want to run queries as if it was one cluster, but insert
data only on the local nodes, and not migrate data between different
locations.
If that's the case, you could also search both clusters and aggregate
the results in your app. Or, you can put all nodes in the same ES
cluster, and use shard allocation to make sure local data gets indexed
in local nodes.
Sorry, that would be "Shard Allocation Filtering". See here:
In case you find out other options, I'd be really interested
?. You code issues the search request, you can decide which one to hit.
On Thursday, March 1, 2012 at 9:02 AM, ajan wrote:
I have a requirement to analyze on how elasticsearch be used in
federation.
Let me try to explain;
Lets say, as a part of a big enterprise, I have a ES cluster - ES001
with 3 nodes, using the default configurations of 1 replica and 2
shards deployed in LA, USA. Another ES cluster - ES002 with the same
configuration as ES001 deployed at Cork, Ireland.
Is there a way to synchronize between these 2 clusters? I'm not
talking about data replication, can a search query on cluster ES001 be
directed to ES002 based on some parameter?
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.