I don't know, but remember that redis writes aren't hitting the disk synchronously.

Writes are saved to an in-memory table first and are burst to disk asynchronously. But, when the burst happens, the writes to disk are sequential - this is key. No reads to the disk are done before writes (which occurs in MySQL).

Same idea in Cassandra: http://wiki.apache.org/cassandra/FAQ#reads_slower_writes

Writes can be faster if you don't have to write through to a slower media; also it would be faster if you can append to the end rather than seek to a specific location. Seeking takes up the majority of the time spent doing disk IO (not talking about SSDs).