Hampton Roads Ruby Users Group (757.rb) February Meetup

Jim Van Fleet plans on comparing and contrasting three different groups, and talking about what kind of problems match the different kinds of technologies. Unlike MySQL and Postgres, for example, which although they have different feature sets, basically do the same thing at the end of the day, the technologies that are being lumped together under the NoSQL flag in many cases have nothing to do with each other:

Document databases
These include Mongo and Couch. Ilya Grigorik includes Tokyo Cabinet in this category, and I’ll mention why I don’t (with an aside about Tokyo’s other benefits).

Hash tables
There are like a zillion of these. Redis is quite popular, memcached was the first. Talking about benefits and genesis is pretty straightforward, but I’ll mention the points of contrast in the ones that I know about.

The Modern Wonders of the World
Amazon’s Dynamo and Google’s BigTable are an inspiration to many implementers of NoSQL technologies. Even those implementers that aren’t directly working on related technologies know about them.

Dynamo is a lot like a distributed hash table with very particular rules and some backend wizardry.

BigTable is an entirely new way of modeling data and “doing an application”.

Cassandra, in particular, is a technology that uses elements of both, and is a major frontier. I can talk a little bit about what the benefits and costs are for investigating Cassandra today.