AI

Fleet management at scale: How Google manages a quarter million computers securely and efficiently

Abstract

Providing Google employees with the tools they need to be productive requires our engineering teams to support a diverse fleet of nearly a quarter-million computers across four operating systems. Operating at this scale requires an efficient approach to imaging, managing, and securing our fleet without hampering user productivity or shifting the maintenance burden to our support staff. This paper discusses how Google has developed reviewable, repeatable, and automated backend processes using flexible open-source systems management software.