Backing up large amount of Git data
At GitHub we store large amounts of Git data and are constantly adding to it. Our old backup solution, based on generic tools, was no longer enough. It was too expensive, slow and required too much attention from our engineers. Over the past year we've developed an alternative solution which is built with Git specifically for efficiently backing up Git and GitHub, allowing us to build a much more reliable system. In this talk we'll walk through the requirements and the design process which lead to the new system.
Carlos has been working with or on Git for the past eight years. He is currently in the Git Infrastructure team at GitHub helping keep the Git data safe and available.