Harnessing the Data from Millions of Connected Cars: GM’s Journey to YugabyteDB
summary
GM migrates the Vehicle Data Factory from Apache Cassandra to YugabyteDB, replacing week‑long node adds with node additions in less than a day and improving consistency versus Cassandra. Bulk migration loads drop from an 8‑hour SLA to under 3 hours — hitting about 3 million writes/sec during tuning (initially ~600–650k TPS). Application latencies shrink roughly 10x (operations that took ~1 minute now complete in <10s). System maintains continuous availability, higher node density and vendor support throughout the cutover.
AI-written summary · about
chapters (10)
- 0:00Introduction
- 2:37Joining Yugabyte and its vision
- 5:47Vehicle Data Factory explained
- 8:15Cassandra challenges at GM
- 10:42Critical database capabilities
- 13:34Why Yugabyte was chosen
- 16:45Migration and performance gains
- 20:43Performance improvements post-migration
- 22:28Support and partnership
- 24:57Takeaways and future plans
related talks
28mDisrupting the Status Quo with Distributed SQL to Deliver Data-Driven Solutions
26mEfforts Toward Adoption and Uncovering Hidden Features
27mSparking our way to YugabyteDB: Our journey from ScyllaDB to
36m5 Hard-Learned Lessons from Successful Database Migration and Modernization
19mEvolve: A Database Journey from Ground to Cloud
31mMake it Painless: YugabyteDB Database Migration Solutions
31mYugabyteDB Voyager: Simplify Data Migration to Accelerate Cloud Adoption
22mYugabyteDB: Discover What's New