If i have a huge database and i want to improve performance , either i can
Increase number of shards
Partition a single index logically (say one index will hold a year's
data) and restrict the search to the required index only. I agree we get
some mileage here as we are restricting the search to the required index
only.
Is there any other advantages for multiple index over multiple shards here ?
I would prefer multiple indices as it can grow dynamically without
reindexing.
When using multiple shards and you want to increase the number of
shards you'll need to reindex the whole index.
If i have a huge database and i want to improve performance , either i can
Increase number of shards
Partition a single index logically (say one index will hold a year's
data) and restrict the search to the required index only. I agree we get
some mileage here as we are restricting the search to the required index
only.
Is there any other advantages for multiple index over multiple shards here ?
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.