You will need to understand the possible range of key values your application will create, and then split those up to balance the load around your cluster.
In general the RandomPartitioner is a good first step. Why are you going with the ByteOrderedPartioner ? Aaron On 27 Jan 2011, at 20:35, Matthew Tovbin wrote: > Hey, > > Can anyone suggest me how to manually generate tokens for Cassandra 0.7.0 > cluster, while ByteOrderedPartitioner is being used? > > Thanks in advance. > > -- > Best regards, > Matthew Tovbin.
