Increasingly slower read operations
This content applies only to On-premises and Client-managed cloud environments
Read operations take a prolonged time, which impacts system performance.
SSTables reduce read capacity because records might be spread across several SSTables, as a result of frequent write operations, modifications, and deletions.
- Check the number of SSTables per column family (table) by using the
nodetool tablestats
command. - Analyze the outcome to determine the root cause of the issue.Depending on the
nodetool tablestats
outcome, you might want to perform such actions as adjusting your compaction throughput, expanding the memory size for file caching, and so on. For more information, see the Apache Cassandra documentation.
Previous topic Input/output blockages in the Cassandra cluster Next topic Unable to connect to Cassandra cluster