Building FinTech Streaming Applications with Pulsar
summary
Apache Pulsar promises one unified streaming/messaging plane that handles petabytes and millions of topics with per-tenant isolation and built-in geo-replication. It runs Kafka/AMQP/MQTT protocols, supports Flink/Trino/Spark integrations, CDC and SQL-on-streams, and tiered storage out to S3 so long retention avoids costly SSDs. Subscription modes (key_shared, exclusive, shared) give exactly-once or fanout semantics, and a production user (BestPay) scales to ~500 million users after consolidating clusters.
AI-written summary · about
chapters (21)
- 0:00FinTech streaming with Pulsar
- 1:06Apache Pulsar overview
- 2:10Multi-tenancy in Pulsar
- 3:12Scalability for FinTech
- 4:36FinTech use cases
- 5:40Async APIs benefits
- 7:08Open source advantages
- 8:13Geo-replication & cloud
- 9:14Real-time analytics in Pulsar
- 10:21Protocol support
- 12:27Subscription types
- 12:49Pulsar vs Kafka
- 14:16Schema & security
- 15:00Best Pay case study
- 15:45Pulsar with Flink & Spark
- 16:48Analytics with Pinot & Trino
- 17:32Migration strategies
- 18:15Learning resources
- 19:41AI & event systems
- 20:25Streaming tech updates
- 21:07Q&A
related talks
51mEvent Streaming and Processing with Apache Pulsar
46mApache Pulsar: Finally an Alternative to Kafka?
45mApache Pulsar: Finally an Alternative to Kafka?
50mApache Pulsar 101: architecture and features
27mLeveraging Event Streaming to Super-Charge your Business Solutions.
31mThe Power of Apache Pulsar. Harnessing Dapr to Build High Scale Messaging
34mBuild ML Enhanced Event Streaming Apps with Python Microservices
42mBuilding resilient and scalable API backends with Apache Pulsar and Spring Reactive