Hi All How performant is M/R on Cassandra when compared to running it on HDFS? Anyone have any numbers they can share? Specifically how much of data the M/R job was run against and what was the throughput etc. Any information would be very helpful.
-- Cheers Bill