About two years ago, with a few others, I embarked on a journey to start a web operations team at Bloomberg. This stemmed from, at the time, the obvious need for a cohesive vision for the design and administration of infrastructure web services; furthermore, Bloomberg simply needed to automate a whole heck of a lot of manual processes surrounding launching a web service.
Up until a little over a year ago this effort was a skunkworks operation, but the proof is in the pudding, and in March 2014 I was given a tiger team of engineers to “automate all the things” for Bloomberg’s web infrastructure. In only a few months time this team was able to make an amazing impact in areas of configuration management and internal software package deployments.
Earlier this year I was given another amazing opportunity to build out Bloomberg’s Platform Engineering team. This new team is primarily responsible for the design, maintenance and administration of Bloomberg’s configuration management infrastructure. This a core bit of infrastructure which is being used to automate our cloud compute environments. That is an intentionally vague mission statement, and it is meant to cover the lifecycle of infrastructure and system deployment.
Over the course of the next few months the Chef community will begin to see the fruits of our labor in the form of some solid cookbooks for buliding reusable infrastructure. We plan to talk about our successes, and failures, implementing Chef at the scale of Bloomberg’s infrastructure.