Brian Stevens, Google Cloud | OpenStack Summit 2017
summary
Argues that hyperscale clouds rely on Linux, open source and Intel and that Google’s stack contributes concrete projects: Spanner’s distributed time synchronization powering global consistency, gRPC as an Apache project for streaming RPC, and TensorFlow open source canvas for ML. Describes operational playbooks turned into CRE (Customer Reliability Engineering) to train SRE-like teams, mentions custom ML ASICs and a global edge footprint in 173 countries and regions, and urges marrying on‑prem and cloud services.
AI-written summary · about
chapters (12)
- 0:00Introduction and welcome
- 0:43Open source and Google's roots
- 2:08Google's open source contributions
- 3:54Developer tools and Kubernetes
- 5:22Enterprise cloud adoption challenges
- 7:27OpenStack vs Google's engineering culture
- 8:53Cloud infrastructure challenges
- 10:17SRE and operations in cloud
- 12:44Data and machine learning focus
- 14:50Hybrid cloud and multi-cloud strategies
- 15:56Edge computing and latency
- 17:41Future of open source in IT
