Osiris: When Big Data Is Too Big for HBase

As the #1 job site in the world, Indeed delivers hundreds of millions of searches per day to job seekers. To give our users the best experience possible, we analyze petabytes of data per day for machine learning, A/B testing, and reporting.

Learn how the Search Quality team at Indeed developed Osiris, a horizontally scalable key-value store built on Hadoop. Osiris is flexible enough to be used in everything from big data analysis to latency-sensitive, user-facing applications.

This talk will cover the requirements and scaling challenges we faced that led to the development of Osiris. We'll also discuss the details of how we built Osiris, including its unique key design and highly configurable storage engine -- both of which allow for use in a wide variety of applications. We'll end this talk with specific examples of how Osiris is used at Indeed.