News for hackers. Stuff that matters.

A comment on The Opposite of a Bloom Filter.

Re:The Opposite of a Bloom Filter by cnlwsu on Monday July 16, 2012 @02:49PM (#4251939)
I had pretty much exact problem. We keep index of row keys in Cassandra on time partitioned data (by hour), but do not want to re-insert the index each time there is a piece of data.

A ConcurrentLinkedHashMap (http://code.google.com/p/concurrentlinkedhashmap/) solved it easily. It does use more RAM though... but an insignificant amount on our heap.

[ Parent | on HN ]
Re:The Opposite of a Bloom Filter by sprsquish on Monday July 16, 2012 @09:27PM (#4253655)
This is almost exactly what our use case is. I slightly altered the implementation to be more GC friendly though: http://squishtech.posterous.com/addendum-to-the-opposite-of-...
[ Parent | on HN ]