PoH – Part 1 – Introduction

By: CriteoLabs / 25 Aug 2014

Last month we released our new fully scalable & distributed learning infrastructure called PoH (Prediction on Hadoop), enabling us to train our prediction models on very large amount of data and providing more accurate predictions.

Have a look back 4 years ago on this amazing story:

These technical evolutions are strongly correlated with Criteo’s growth, as we become bigger and bigger every day, we always have to anticipate this growth and take fast and smart decisions, combining short term vision (we can’t stop growth for technical reasons) and long term (we want to keep staying ahead our competitors with cleaner infrastructure and smarter algorithms).

Today, we have identified some potential limits of our solution which could be reached in the next 6 months, and we are already working on improvements by testing new technologies and algorithms.

Author: Laurent Vion

  • CriteoLabs

    Our lovely Community Manager / Event Manager is updating you about what's happening at Criteo Labs.