I have been doing some research on adding shards and replicas after they
have been created. It appears that replicas are no big deal. Use the API,
and elasticsearch will distribute copies evenly across your cluster.
In one article I read on nabble, adding a shard after the fact
causes repartitioning and is expensive and it is better to create another
index. I'm not quite sure if I follow along completely. Could someone
elaborate?
The idea of creating another index means that you can work around the limit
of not being able to change the number of shards. That is possible since you
can search across indices.
I have been doing some research on adding shards and replicas after they
have been created. It appears that replicas are no big deal. Use the API,
and elasticsearch will distribute copies evenly across your cluster.
In one article I read on nabble, adding a shard after the fact
causes repartitioning and is expensive and it is better to create another
index. I'm not quite sure if I follow along completely. Could someone
elaborate?
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.