Case Study: Palo Alto Networks boosts stream processing performance and cuts operational complexity with ScyllaDB

How Palo Alto Networks Replaced Kafka with ScyllaDB for Stream Processing

The cybersecurity firm Palo Alto Networks faced the challenge of correlating millions of disparate network security events per second into coherent stories to identify threats. This required a system to process terabytes of data daily with zero tolerance for loss, all while managing thousands of unique customer deployments efficiently and cost-effectively. Their initial architecture required both a database and a message queue like Kafka, which introduced high operational overhead and complexity.

By leveraging their existing ScyllaDB NoSQL database as both a data store and a message queue, Palo Alto Networks implemented a streamlined solution. This approach from ScyllaDB enabled them to completely eliminate Kafka from their deployment. The results were a significant reduction in operational cost and complexity, alongside a measurable increase in system performance.

Palo Alto Networks

Daniel Belenky

Principal Software Engineer