The function you are using I think can result in certain slots having up to 10x more keys then others. It may only be apparent in clusters with lots of keys. It would be nice to be able to choose a partitioning function that closely met our use case.
Comment From: hpatro
@jonathandelanders if you're still following the project.
Do you have any particular dataset/use case to share where such hot slot occurs? I was also thinking it might be beneficial for users to choose the slot partition algorithm.
E.g.: Cassandra provides option to choose the data partitioner https://docs.datastax.com/en/cassandra-oss/3.0/cassandra/architecture/archPartitionerAbout.html