I am trying to set the replication to async. I tried the following.
$ curl -XPOST 'http://localhost:9200/myindex' -d '{
"settings": {
"replication": "async"
}
}'
Is it the correct way to do it?
Thanks for the reply. Yes, In our user case, we only want eventual
consistence in our data across the replicas. Will setting replication to
async increase write performance? Thanks again for helping out.
On Fri, Jul 4, 2014 at 12:06 PM, David Pilato david@pilato.fr wrote:
--
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 elasticsearch+unsubscribe@googlegroups.com.
The default setting of Elasticsearch is replication mode "sync" for good
reason, this ensures the replica quorum (which is often determined as
"eventual consistency" but this is really something more because it
describes properties of the "last-seen" value for update).
Replication setting of "async" does not increase write performance on the
cluster, and it also does not enable eventual consistency. In fact it
disables the client's visibility of consistency. Replication setting to
"async" means an early return to the client who is listening for responses.
The responses under "async" mode return immediately after primary shard
operations but before any actions on replica shards have been executed. The
client can proceed earlier so this may help at the client side if
the client wants to finish the indexing process sooner regardless if
replica quorum has been reached at that time or not.
Thanks for the reply. Yes, In our user case, we only want eventual
consistence in our data across the replicas. Will setting replication to
async increase write performance? Thanks again for helping out.
On Fri, Jul 4, 2014 at 12:06 PM, David Pilato david@pilato.fr wrote:
--
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 elasticsearch+unsubscribe@googlegroups.com.
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.