Cassandra powered workflows to automate at scale
summary
Demonstration shows Apache Cassandra used as the primary workflow store with the open source Automatico toolkit, creating per-workflow Cassandra tables (instance id + blob state) and event-published denormalized views for reporter/support queries. A support-incident workflow runs end-to-end: create instance, add comments, auto-close on keyword, priority routing for “Apache” company, and UI-backed queries — all working against a geo-replicated cluster. Caveats: tombstone buildup from deletions, secondary-index performance at scale, and blob size limits for large document payloads.
AI-written summary · about
chapters (17)
- 0:00Introduction
- 0:30About me
- 1:12What are workflows
- 4:40Workload as a Service
- 6:47Workload of Function
- 7:56Workflow as Function Flow
- 9:21Stateful vs Stateless
- 10:41Apache Cassandra
- 12:26What is to be stored
- 14:20What happens behind the scenes
- 16:22Implementation
- 17:27Demo
- 18:06Requirements
- 19:04Annotation
- 22:06Project overview
- 23:46Project demo
- 31:45Recap
related talks
46mApache Cassandra: 8 years in production
53mApache Cassandra. Inception - all you need to know
26mUnleash the Power of Apache Cassandra: Scalable Recommendation Systems, Real-time Fraud Detection
32mMigrating Workloads from Mainframes to Cassandra - Problems, Experience and Learnings
27mDesigning Modern Data Platforms with Apache Cassandra and Beyond
40mApache Cassandra in High-Performance Systems Development
32mCassandra at Yammer #bbuzz
29mCassandra - an introduction