Event Streaming for the Best of All Worlds
summary
Apache Pulsar presents a cloud‑native event platform that separates compute and storage, delegates durable log storage to Apache BookKeeper, and uses ZooKeeper for cluster metadata; it supports both pub‑sub and queueing models, tiered cold storage, serverless Pulsar Functions and many connectors. Pulsar claims easier horizontal scaling and built‑in geo‑replication vs Kafka, had more contributors than Kafka by 2021, and is available as open source or managed services (demo shows tenant/topic creation working end‑to‑end).
AI-written summary · about
chapters (27)
- 0:00Introduction and welcome
- 2:07Speaker background
- 3:37Event streaming overview
- 4:53Defining events
- 6:27Asynchronous communication
- 7:34Event streaming explained
- 8:50Complex event processing
- 10:13Messaging vs streaming
- 12:22Batch vs streaming
- 14:06Pulsar messaging patterns
- 15:34Why event streaming matters
- 18:38Introduction to Apache Pulsar
- 20:30Architecture design
- 22:35Pulsar functions and connectors
- 24:43Pulsar ecosystem and adoption
- 26:15Pulsar components
- 27:41Scalability and persistence
- 28:51Pulsar vs Kafka performance
- 30:34Key differentiators
- 32:21Geo-replication details
- 34:10Multi-tenancy explained
- 35:05Message subscription modes
- 36:20When to choose Pulsar
- 37:33DataStax offerings
- 38:56Kafka limitations
- 40:53Astra Streaming demo
- 44:46Resources and conclusion
related talks
53mEvent Messaging and Streaming with Apache Pulsar
48mApache Pulsar: Event Streaming in The Cloud Native World
28mBuild your next-generation stream platform based on Apache Pulsar
45mScalable distributed messaging&streaming with Apache Pulsar
45mApache Pulsar: Finally an Alternative to Kafka?
48mBest of Both Worlds: Apache Pulsar and Apache Kafka
46mScalable messaging and streaming with Apache Pulsar
51mApache Pulsar 101: architecture, concepts & benchmarks