We have around 5000 Nodes in our setup.
If I distribute the Nodes across 30 different Clusters will the cross cluster search performance be significantly be faster as compared to if I have nodes distributed across 100 smaller Clusters?
I do understand that the network trip will increase in case of 100 Clusters as compared to 30 clusters.
Assuming that the Node , Index and Shard configuration remains the same in both the case, Is there any other factor which might impact the performance.
We are concerned about the search response time. That is the biggest criteria for us.
Also ,Will the slowest cluster(The Hardware is pretty slow) decide the performance on the search in CCS elastic?
The only way to be sure is to try it, the answer is going to depend on so many details specific to your setup & workload. But I would not expect it to make a significant difference to performance in most cases.
This answer also depends on details specific to your setup & workload, but I would expect it to be approximately "yes".
Hi @DavidTurner , Thanks for the answer. If we do targeted Index search (Where the Indices which are searched are present in the Faster Machine)in the Multi Cluster environment, in that case, if an Index is not present in a slower Cluster, the search performance will not get slow I believe.
What's your take on this?
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.