discussion · Code of Architecture

White paper: Large-scale cluster management at Google with Borg

Name: White paper: Large-scale cluster management at Google with Borg · Code of Architecture
Uploaded: 2023-08-25
Duration: 48 min 19 s
Description: Recording of a Code of Architecture discussion with Alexander Polomodov and Salikh Fakhrutdinov on Google's white paper Large-scale cluster management at Google with Borg, cluster management, SRE, and platform engineering.

August 25, 2023 · Code of Architecture · IT Fest Kazan

A special Code of Architecture episode from IT Fest in Kazan: a discussion of Google's Borg white paper, covering large-scale cluster management, workload scheduling, resource utilization, reliability, and the engineering practices behind modern infrastructure platforms.

Watch on YouTube White paper

Participants

Alexander Polomodov
Salikh Fakhrutdinov · Senior SRE · Tinkoff Origination Platform

Discussion focus

01Why did Google need a unified cluster management system at Borg scale?
02How does Borg place workload, manage resources, and improve infrastructure utilization?
03Which reliability, resilience, and self-healing mechanisms are built into the platform?
04How are SRE practices connected with the architecture of a cluster manager?
05Which Borg ideas influenced Kubernetes and modern platform teams?
06What parts of Google's approach can be transferred to large product companies?