Hadoop is open-source software for reliable, scalable, distributed computing. This talk covers the distributed processing of large security data sets across clusters of computers using simple programming models. The system is designed to scale from small environments to clusters of thousands of machines each providing local computation and storage. High availability is achieved through software enabling cost-effective scaling and resilience. We show wow to process terabytes of petabytes of security data using the same platform as Yahoo!, Facebook, EBay, LinkedIn, Last.fm, Ning, Microsoft, Quantcast, Spadac, Twitter, Tegatai and countless other firms.