Tempest is OpenStack’s integrated test suite which aims to provide validation that OpenStack is working. As such it is run as a gating on job on all proposed commits to OpenStack. It is designed to run against an operational OpenStack cloud, which includes everything from a devstack deployment to a public cloud. Tempest originally started as just a small number of integration tests to verify that the various OpenStack projects worked together. It has since grown into one of the top 5 most active OpenStack projects with several different classes of testing and validation. This talk will provide an overview of what Tempest is and how it works. Providing an explanation of the philosophy behind the project, and insight into why things are setup a certain way. Additionally, it will cover some of the features in tempest and how to configure and run it locally with or without devstack. The most immediate “”killer apps”” that the Swift+ZeroVM combination offers are in data processing: – oil, geo, mining – TV and movies – photo and picture processing Some less obvious but even more interesting are things like the possibility to completely redesign an SQL DB and create a structured query language that processes binary unstructured data instead of rows and columns.