We are asking this for guidance on high performing schema design with secondary indexes for reads. We have 7 columns in a column family. Our queries need to retrieve rows by setting where clause for 6 of the columns. Due to this reason we are adding secondary indexes for these 6 columns. We are noticing that as we increase the "where" clauses in cassandra-cli client the performance degrades. We checked this behavior with "Astyanax" java client.
Can you please suggest how we go about increasing the read performance.
Performance of Cassandra with multiple secondary indexes
(2 posts) (2 voices)- Started 9 months ago by rprvnreddy
- Latest reply from xedin
- This topic is not resolved
-
rprvnreddy
MemberPosted 9 months ago # -
The only option I see would be to try to build a composite column of those 6 and query it instead of secondary indices, this should be a good start http://www.datastax.com/dev/blog/introduction-to-composite-columns-part-1
Posted 9 months ago #
