100 Terabytes, 5 Billion Documents, 10 Billion Parameters, 1 Billion Inserts/s

parameter server
distributed learning
Published

December 15, 2012

We’ve been busy building the next generation of a Parameter Server and it’s finally ready. Check out the OSDI 2014 paper by Li et al.; It’s quite different from our previous designs, the main improvements being fault tolerance and self repair, a much improved network protocol, flexible consistency models, and a much more general interface.

In the next few posts I’ll explain the engineering decisions that went into this system which is capable of solving problems as diverse as very high throughput sketching, topic models, and optimization. And yes, it is open source.