I am trying to add a huge dataset into cassandra using phpcassa. The dataset comprises of our site events over the last few years. I have a super column family to which I want to append events. My current approach is to get the existing data from cassandra, append the new events and then write it back. The approach does not seem to scale beyond a few 100k records. What's the best way to append data to a super column family?
Just append the new events as new supercolumns, there's no reason to read-and-rewrite the old ones.
See http://www.datastax.com/docs/0.8/data_model/supercolumns for more on supercolumns.