i looking probabilistic counting , set membership using structures such bloom filters , hyperloglog.
is there support using such data structures , performing operations on them atomically on server-side, through user-defined functions or similar? or way me add extensions such functionality?
(i ingest data through system , batch updates reduce contention, far simpler if handled in database server.)
you have implement them client side. common approach every x min serialize/insert hll keep in memory on system , merge them on reads across interested range (maybe using rrd type approach different periods beyond x min). not durable, depending on usecase might mean more complex.
although seems close fit c* think 1 of big issues deletes, can work around them. theres proof of concept c* side implementation here:
http://vilkeliskis.com/blog/2013/12/28/hacking_cassandra.html
that can working "well enough". https://issues.apache.org/jira/browse/cassandra-8861 may watch.