BDM39: How MindGeek's Ad Network uses Big Data Technologies to push Billions of Impressions per Day
MindGeek's Ad Network pushes Ads to some of the biggest sites on the Internet. This produces a massive amount of data that must be analysed in real time to adjust bidding patterns, detect fraud, debug issues, and bill customers. To do so, we use standard Open Source technologies such as Kafka, Samza, Hive, etc. In this talk, we'll present our technical architecture and how it's being used by our different teams (Data Science, Sales, Monetization and Fraud Detection) at MindGeek.
(by Olivier H. Beauchesne, Lead Data Scientist at MindGeek)