I was just wondering how searches are performed. I guess it's a kind of map
reduce where either each shard or each node is queried and then the
initiating node collates.
So what happens if for a particular index, you have more nodes than shards?
Say you have 3 nodes and 2 shards. Does a search operation still create 3
requests - one for each node - or does it only query the nodes with shards?
On Saturday, June 4, 2011 at 5:13 PM, Paul Loy wrote:
Hi Shay,
I was just wondering how searches are performed. I guess it's a kind of map reduce where either each shard or each node is queried and then the initiating node collates.
So what happens if for a particular index, you have more nodes than shards? Say you have 3 nodes and 2 shards. Does a search operation still create 3 requests - one for each node - or does it only query the nodes with shards?
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.