Complex Event Processing Made Easy with Cassandra and Quine Streaming Graph
summary
Quine builds a property graph from incoming JSON logs using Cypher-like queries, runs standing queries to emit pattern matches in real time, and persists hot and cold state to Cassandra (or Astra) so matches can influence future ingestion. Demo detects a password‑spray pattern and flags attacker/victim; single-host throughput ~9–10k events/sec, cluster runs near 100M events/sec in their benchmark, and typical round‑trip alert latency measures ~13 ms. actor‑oriented nodes and deterministic node IDs enable horizontal scaling.
AI-written summary · about
chapters (9)
- 0:00Welcome
- 1:05Stream vs complex event processing
- 3:54Stream vs complex event processing differences
- 9:00Quine streaming graph
- 10:49Password spraying attack detection
- 16:53Cassandra for CEP scaling
- 21:08Live demo password spraying attack
- 25:01Quine and Cassandra in production
- 27:28Q&A
related talks
53mQuerying Events
34mChasing the Stream Processing Utopia
41mStream Processor as a Database: Building Online Applications
29mEnhancing Vector Databases with Real-time Stream Processing
38mPostmortem: Evolution of an Event recording solution
57mLessons in Building a System that Processes More than 70 Billion Events Daily
57mTaming IoT Data: Making Sense of Sensors with SQL Streaming
38mHow to handle 100k events per second