Hacker News new | ask | show | jobs
by glidek 3435 days ago
> Having a large partition also means the data in it cannot be distributed around the cluster.

Why can't a large partition be distributed around the cluster?

1 comments

Cassandra uses consistent hashing. A partition is a segment of data identified by the partition key to determine which node in the consistent hash ring owns that data.

You cant break down partitions any further because it's just a name for the smallest cohesive set of data owned by a hash key, so instead it's advisable to use more partitions with data modeling rather than making them huge.