Spark reading from cassandra

[Amit Sharma]

Hi, i have a question while we are reading from cassandra should we use
partition key only in where clause from performance perspective or it does
not matter from spark perspective because it always allows filtering.


Thanks
Amit