Please advise me with steps to troubleshoot slow read performance in dse 6.7.4 with search enabled feature.
i am checking with read cassandra_table_readlatency_95thpercentile and i can see latency is about 9 ms which was 500-600 us before.
would 9ms will affect performance or is it normal.
There is no "universal" latency value that applies to all clusters. Each cluster is unique and the idea of "normal" is whatever it is for your specific cluster.
The reason for this is that there are a lot of moving parts and no two clusters are the same -- hardware configuration, use case, data model, access patterns, cluster load, disk throughput, network throughput, etc.
If you think there is a problem with your cluster or if you need support assistance, please log a ticket with DataStax Support. Cheers!
Have you check on https://academy.datastax.com/#/online-courses/ there is a session in DS310 course on DSE Search on "Trouble Shooting Slow Queries"
5 People are following this question.
