Superfast event stream processing with a converged database

on YouTube ↗31m

" summary

Oracle’s converged database uses Oracle GoldenGate for parallel capture and a Memoptimized Rowstore for burst ingest, claiming 25 million rows per second and even “more than 21 trillion inserts per day on a single two‑socket server.” Partitioning, RAC and sharding split workload; in‑memory columnar scans run analytics at billions of rows/sec; automatic data optimization plus DBMS_SCHEDULER handles rollups. A demo replaces Prometheus: 100 hosts sending 150,000 metrics every 15s, Grafana queries run PromQL translated to SQL and alerts fire from DBMS jobs, proving the pipeline works end‑to‑end.

AI-written summary · about

" chapters (16)

1. 0:00Intro
2. 0:27What is Event Data?
3. 1:11Properties of Event Data
4. 2:01What is Event Stream Processing?
5. 3:42Event Stream Processing Architecture Choices
6. 5:41JSON in Oracle Ideal for Event Stream Processing
7. 6:37Oracle Golden Gate for Fast Data Capture Leading real-time data capture engine, available as a fully managed Service
8. 7:48Memoptimized Rowstore
9. 8:57Partitioning Paralelism and Elliciency for Event Stream Processing
10. 10:03Real Application Clusters (RAC) Industry-Leading Scale-Out For Event Stream Processing
11. 15:35Attribute Clustering Data organization to maximize event analytics performance
12. 16:52SQL for Event Stream Processing
13. 19:17Oracle Machine Learning for Event Streams
14. 20:51Automatic Data Optimization and Database Scheduler
15. 23:16Downsampling Event Streams Using Database Scheduler
16. 28:35Event Streams Demo Setup A DevOps Deployment

" related talks

• 20mOracle stream analytics on data in motionAlex Kotopoulis · Oracle CloudWorld 2022
• 10mBuilding High-Performance Financial Data PipelinesAnandan Dhanaraj · Conf42 Machine Learning 2025
• 13mOracle Globally Distributed DatabaseShailesh Dwivedi · Oracle DatabaseWorld: Multicloud AI Edition 2025
• 30mCase Study Productizing AsyncAPI for Data Replication, Changed Data CaptureVeeratteshwaran Sridhar · AACoT 2023 Bangalore
• 28mBest of Stream + Batch: When Streaming Analytics Isn’tRohan Garg · DataEngBytes 2022
• 24mOracle's vision for simplifying data-driven apps and analyticsJuan Loaiza · Oracle DatabaseWorld 2022
• 30mReal-Time Data Ingestions for Data Fabric and AI ApplicationsDeniz Sendil · Big Data Conference Europe 2025
• 48mContinuous Optimization for distributed BigData analysisKai Sasaki · HighLoad++ Moscow 2018