We are importing data from one column family into a second column family via "nodetool refresh" but we are finding a secondary index is performing slow and the machine CPU is pretty much idle. We are trying to bulk load data as fast as possible.
Are secondary indexes concurrent or single threaded?
