Google Cloud Spanner: Going global with relational database
summary
Cloud Spanner is a fully managed relational DB that provides synchronous replication across three availability zones, automatic splits and rebalancing of key ranges, and online scaling of logical nodes with no downtime. It uses Google’s true time (atomic clock + GPS) to give globally consistent timestamps, supports SQL2011 for reads while writes use client libraries, and recommends one node ≈ 10k reads / 2k writes / 2 TB (production → 3 nodes). Demo shows console UI, monitoring, and Quizlet tests show Spanner keeps latency stable as load grows while MySQL latency spikes.
AI-written summary · about
chapters (8)
- 0:00Introduction
- 1:50Why Cloud Spanner exists
- 8:39How Spanner works
- 16:27Query execution in Spanner
- 23:11Schema design best practices
- 26:23When to use Cloud Spanner
- 29:17Cloud Spanner demo
- 35:03Summary and next steps
related talks
47mGoogle Cloud Spanner: Global consistency at scale
33mGoogle Cloud Spanner: Global consistency at scale
41mSpanner - Google's Distributed Database
49mSpanner - fully managed horizontally scalable relational DB with ACID transactions
24mnative databases and storage for games
01h 02mDiamond Sponsor Hall
29mBuild an Application at Google Scale with Spanner, Cloud Foundry, and Google Cloud Platform
01h 08mKeynote: Building Spanner