White Paper: Ooyala

Title: Designing a Scalable Database for Online Video Analytics

Description: A deep understanding of how video content is consumed is critical to maximizing revenue and viewership for online video content. Publishers are increasingly looking for sophisticated ways to monetize and syndicate their video to ensure they are reaching the largest number of viewers and maximizing the value of their online video content. Video analytics is the key to unlocking the full potential of online video content, and leveraging the proper technology and infrastructure to manage the vast amounts of viewership data is an important component for an online video analytics platform.

There are a vast number of possible databases to choose from to store viewership data, ranging from the more traditional relational databases (such as MySQL) to distributed databases (like Cassandra and HBase). Traditional relational databases fall short when confronted with the terabytes of data generated from online video. Forcing a relational database model on these large-scale datastore problems often results in sharding architectures that are difficult to maintain and scale. Fortunately, there are alternate solutions designed to more effectively handle large amounts of loosely structured data like data generated from online video viewership.