Oracle Globally Distributed Database
summary
Oracle describes a single logical database sharded across regions that hits 1 million transactions per second for one customer, stores petabytes on 104 commodity servers (5,408 cores, 77.4 TB RAM), and offers online rebalancing with no downtime. Replication options include Raft-based replication for sub‑3‑second failover and zero data loss, Active Data Guard redo-level replicas, and GoldenGate SQL replication; supports vector search (23ai), multi‑cloud, and autonomous operation.
AI-written summary · about
chapters (17)
- 0:00Introduction to distribution
- 0:27Architecture and use cases
- 2:21Data handling mechanisms
- 3:01Customer sovereignty examples
- 3:40Banking regulation compliance
- 4:12Scaling and performance
- 4:56Scalability and reliability
- 5:15High-performance analytics
- 5:40Never-down payment processing
- 5:52Data distribution methods
- 6:09Advanced sharding techniques
- 6:55Replication and survivability
- 8:49Replication alternatives
- 9:19AI integration and natural language
- 10:47Autonomous and cloud services
- 11:43Summary and key takeaways
- 12:55Closing remarks
related talks
23mSharding for hyperscale and geographically distributed relational databases
32mSharding for mission-critical, internet-scale apps
19mBuild Global-Scale Apps Using Globally Distributed Autonomous Databases
30mOCI GoldenGate: Elastic, real-time data integration
34mOracle Autonomous Database: simplify your database management
27mAvailability and Performance Tradeoffs in Global Database Deployments
24mRethinking Database Architecture to Embrace Modern App Dev
27mScaling for Intelligence: Distributed SQL Databases in the Age of AI