On Mon, May 31, 2010 at 3:42 PM, Jake Mannix <[email protected]> wrote:
> Voldemort to store rows > of the ItemSimilarity matrix and the user-item preferences matrix, > computing recommendations for an item-based recommender on the > fly by doing a get() based on all the items a user had rated and then > a multiGet() based on all the keys returned, ... > ... > What do you guys think? > Been there. Done that. Works great. In fact, for quite a while, we just used LightHttpd to serve these vectors and didn't even have a multi-get. That still worked well. I agree with Sean that SQL shouldn't be dropped ... this is just an addition.
