Seminarium: Systemy Rozproszone
7 stycznia 2016, godzina 12:15, sala 4070
Paweł Krawczyk

Large-scale cluster management at Google with Borg

Google’s Borg system is a cluster manager that runs hundreds of thousands of jobs, from many thousands of different applications, across a number of clusters each with up to tens of thousands of machines. It achieves high utilization by combining efficient task-packing and machine sharing with process-level performance isolation.

I will present a summary of Borg system architecture and features, desing decisions and analysis of some of its policy decisions. I will also recount lessons learned from operating Borg in production and describe how these observations have been leveraged in designing Kubernetes.

Paweł Krawczyk

Bibliografia:

• Abhishek Verma, Luis Pedrosa, Madhukar Korupolu, David Oppenheimer, Eric Tune, John Wilkes. Large-scale cluster management at Google with Borg
• David Oppenheimer. Borg: The Predecessor to Kubernetes