Brandon Butler writes "Netflix, yes the video rental company Netflix, is changing the cloud game. During the past two years the company has pulled back the curtains through its Netflix OSS program to provide a behind-the-scenes look into how it runs one of the largest deployments of Amazon Web Services cloud-based resources. In doing so, the company is creating tools that can be used by both entire business-size scale cloud deployments and even smaller test environments. The Simian Army, for example randomly kills off VMs or entire availability zones in Amazon's cloud to test fault tolerance, Asgard is a cloud resource dashboard and Lipstick on (Apache) Pig, is a data visualization tool for the Hadoop program; there are dozens of others that help deploy, manage and monitor the tens of thousands of VM instances the company company can be running at any single time. Netflix is also creating a cadre of developers who are experts in managing cloud deployments, and already its former employees are popping up at other companies to bring their expertise on how to run a large-scale cloud resources. Meanwhile, Netflix does this all in AWS's cloud, which raises some questions of how good of a job it's actually doing when it can be massively impacted by cloud outages, such as the one on Christmas Eve last year that brought down Netflix's services but, interestingly, not Amazon's own video streaming system, which is a competitor to the company."