TPC-C Benchmark: Scaling YugabyteDB to 150,000 Warehouses
summary
YugabyteDB runs the TPC‑C benchmark at 150,000 warehouses on a 75-node universe (72 tservers, 3 masters) hosted on AWS c5d. The cluster loads 49.5 TB replicated data, uses 60 client nodes, and simulates ~1.5 million terminals; the run achieves 12.76 million TPMC with >99% efficiency, average transaction latency ~142 ms, CPU ~80%, and low retry rates after two retry attempts, showing the distributed SQL cluster scales and sustains throughput.
AI-written summary · about
chapters (6)
- 0:00TPC-C benchmark intro
- 8:10TPC-C performance challenges
- 9:57Cluster setup for 150k warehouses
- 16:47TPC-C results and metrics
- 23:41TPC-C analysis insights
- 27:17Q&A
related talks
30mFrom monolith to petabyte scale: Built a cloud-native, AI-ready data layer
28mDisrupting the Status Quo with Distributed SQL to Deliver Data-Driven Solutions
30mBuilding Scalable Spring Applications with YugabyteDB
57mNo Downtime, No Limits
26mA Winning Multi Cloud Data Strategy
26mLeveraging a Distributed SQL Database at Tokopedia: Best Practices & Lessons Learned
29mEnterprise Scale: How a Distributed Architecture Can Deliver Scale and Resilience
26mTip and Tricks to Deploying YugabyteDB in a Multi-Cloud World