Enterprises are Increasingly demanding realtime analytics and insights to power use cases like personalization, monitoring and marketing. In this talk we will present Pulsar, a realtime streaming system which can scale to millions of events per second with high availability and 4GL language support. Pulsar provides realtime sessionization, multi-dimensional metrics aggregation over time windows and custom stream creation through data enrichment, stateful processing and filtering using SQL-like language.

We will discuss how Pulsar is being used at eBay, the features and architecture of Pulsar. Topics include:

• Pulsar Real-time analytics pipeline
• Pulsar architecture to support high scalability and availability
• Pulsar event processing framework and language
• Integration of Pulsar with Kafka, Druid and session store

Key Takeaways:

• real-time analytic and processing concepts
• stream processing architecture supporting high availability and scalability
• stream processing framework and language
• integration of real-time stream processing system with Kafka, Druid and session stores