One platform for all: real-time, near-real-time, and offline video analytics on Spark - Databricks

One platform for all: real-time, near-real-time, and offline video analytics on Spark

Download Slides

At Conviva we need to navigate the massive diversity that is the video delivery pipeline in order to help deliver high quality video experiences to every viewer. We have developed a real time decision making engine built on top of Spark to help deliver on that goal. The talk will cover in some detail how Spark allowed us to solve a realtime regression problem to predict and intelligently control video traffic over the internet, and how it is now becoming the backbone of our realtime data processing stack.