The F# community member John Liao has blogged about using the Riak distributed database system with F#.

From Wikipedia: Riak is a NoSQL database implementing the principles from Amazon's Dynamo paper.

Lately, I have been reading the book  Signals and Noise by Nate Silver.  His book references an IBM  webpage that claims the world is creating 2.5 quintillion (1018) bytes of data a day and that 90% of the data that currently exists the world was created in the past two years.  Combine this fact with the rise in  multicore processors,would explain the surging interest in functional programming and NoSQL and MapReduce frameworks.  There is just too much data to be handled by a single machine alone

Continuing the journey in the book Seven Databases in Seven Weeks as I explore MapReduce. MapReduce framework is popularized by Google's seminal paper, but the genesis of that framework started with ideas in functional programming languages.  After all, if functional programming is intended to help one run code on multiple processors on a single server, it's a logical conclusion to extend that computing from multiple processors on a single server to multiple servers.  However, it's the recent technology factors and the data tsunami that I believe encourages people to look at functional programming and NoSQL databases.

What I would like to see in the future is that MapReduce capability is integrated with functional programming in such a way that calling a map/reduce function on a typed collection that recognizes that the typed collection is backed by a NoSQL database that it seamlessly spawns the function to the NoSQL stores in a MapReduce fashion.

read more here