> Is there any way of updating the TTL without being in need of rewriting the > data back again?
No, there isn't. > If not, is running a Map/Reduce task on the whole data set the "best" option If the TTL is made rather infrequently and on a large percentage of the data, which seems to be your case, then that's likely the option I would use. > My last question is regarding to a long term TTL, does this have any > negative impact on the cluster? The main drawback of TTL is that each column has a 8 bytes overhead (a tad more details here: http://www.datastax.com/dev/blog/whats-new-cassandra-07-expiring-columns). It's not a big deal imo, especially now that we have compression, but it's worth knowing. There is nothing specific to having "long term" TTL in any case. -- Sylvain
