Column filters in Cassandra/Pycassa

Question

I am relatively new to columnar database, please forgive ignorance. Lets say I have 1,000,000 columns. I would like to return a random sample of 10% of those columns (ie c0, c10, c20...c999,980, c999,990)

In HBase they have column filters, I could write a column filter that returned every tenth result. Can I do this in Pycassa/Cassanda?

Thank you

Answer 1

The only thing you can do server side is slices. So you can read starting at column=C10 limit=10 to get columns 10-19. Or you can ask for specific columns, so you could ask for every 10th column manually if you knew how many columns there were.

Answer 2

You could do this easily client-side with Pycassa, but Cassandra does not support server-side filtering.