hey List, I consider a way that can read all data from a column family, the following is my thoughts:
1. make a snapshot for all nodes at the same time with a special column family in a cluster, 2. copy these sstables to local disk from cassandra nodes. 3. compact these sstables to a single one, 4. parse the sstable to each rows. My problem is the step2, assume that the replication factor is 3, then I need to copy the data size is: (3 * number of bytes for all rows with this column family), is there any proposals on this? -- *Rick Dong *
