Streaming Analytics with Apache Pulsar
summary
Apache Pulsar is a Kubernetes-native messaging and streaming platform that claims millions of messages a second throughput, uses BookKeeper and ZooKeeper for storage/metadata, and offers tiered storage to offload older data to S3. Aerospike integrates bidirectionally: an XDR connector pushes changes into Pulsar and pulsar-sync writes stream results back into Aerospike for low-latency serving. Users report massive scale (Tencent >10 billion msgs/day, Splunk billions/day) and one customer consolidates RabbitMQ instances into Pulsar, cutting ops costs by roughly 50%.
AI-written summary · about
chapters (10)
- 0:00Introduction
- 1:28About StreamNative
- 1:50Modern Data Needs
- 6:10Apache Pulsar
- 17:06Pulsar and Aerospike
- 21:37Conclusion
- 22:37Questions
- 23:55Streaming vs Messaging
- 29:20Connectivity
- 30:45Summary
related talks
46mApache Pulsar: Finally an Alternative to Kafka?
51mEvent Streaming and Processing with Apache Pulsar
10mA tour of Apache Pulsar
54mApache Pulsar Development 101
50mEvent-Driven Change Data Capture using Apache Pulsar
43mBuilding a Real-time Analytics Application with Apache Pulsar and Apache Pinot
30mFlink + Pulsar: The Path To Unified Batch and Streaming
32mBuilding a Real-time Ingestion Platform using Kafka and Aerospike
