i have a cluster which contains two CFs.  One is a bunch of rows with 10-15 columns per row.  the other is an index of those items with only a few rows, but thousands of columns per row.  i am noticing that the replication of data between the nodes in the cluster is causing a lot of memtable flushing for the index CF.  is this normal?  my memtable thresholds are 128M or 1.5m objects.  Some of the rows in the index CF have 500k+ columns.  Also, this doesn't seem to happen in my dev environment which has a single cassandra node.  It also slows down the rate of flush/compact cycles when there is only one node up in the prod cluster.

Lee Parker