Building a distributed mindset: Taking self-hosted CockroachDB to production
summary
A Procter & Gamble IoT team replaces many regional NoSQL stores with a self-hosted CockroachDB cluster distributed across multiple regions, eliminating the “data graveyard” and giving predictable SQL validation for analytics. Engineering chooses table/row locality and a possible regional-by-row strategy, finds a real performance bug where client-generated UUIDs caused cross-region latency (hundreds ms vs expected tens), and standardizes on Cockroach University plus a local dev flow that spins a Cockroach image into Kubernetes via Tilt and make for reproducible testing.
AI-written summary · about
chapters (7)
- 0:00Welcome & Introduction
- 1:43IoT needs
- 2:58Issues with a NoSQL stack
- 5:18A self-hosted CockroachDB cluster distributed across multiple regions
- 6:11CockroachDB architectural diagram
- 6:43How to grow a distributed mindset
- 12:18The local developer setup
related talks
48mBuilding mission-critical applications with CockroachDB
31mFuture-proofing operational data at scale
29mHow City Storage Systems designed its Data Infrastructure to minimize toil
14mHow CockroachDB can enable Multi-Region Deployments and Data Locality 
23mWhy Squarespace adopted CockroachDB for better global performance at scale
24mNavigating Distributed Databases: Challenges & Opportunities
02mRoachFest 24 Menlo Park | A CockroachDB Conference
15mJourney from PostgreSQL to CockroachDB: Scaling Cisco AI Defense