I am reading a table from cassandra , and I am facing delay in read request. My observation is that there is very big partition in cassandra but when I am running cfstats command it returns
There isn't an easy way of finding the largest partition in a table. That functionality requires a full table scan and that operation isn't allowed in Cassandra.
Have a look at cassandra-sstable-tools. I haven't tried it myself but you might find something there that meets your needs. Cheers!
In addition to what Erick said, you might also look into Spark partition sizes in Spark UI. You won't get exact Cassandra partition sizes there, as Spark partition usually contains many Cassandra partitions.
6 People are following this question.
