Facebook outlined its new state-of-the art system, named Robotron, for managing the tens of thousands of network devices connecting hundreds of thousands of servers globally at Facebook.
In a newly published whitepaper, the team behind Robotron describe key learnings in designing, implementing and operating the Robotron system on Facebook’s production network of data centers, global backbone, and edge point of presence (POPs) over the last eight years.
Robotron, which is an evolutionary leap beyond the automated scripts used to manage the Facebook infrastructure in the early days, now covers multiple stages of the network management life cycle: network design, config generation, deployment, and monitoring. At heart of Robotron is FBNet — a central repository for information, implemented as an object store, where each network component is modeled as an object.
https://code.facebook.com/posts/529305410613534/robotron-top-down-network-management-at-scale/
