Protecting your web applications built on Kubernetes

Tomer Rozentvaig, who leads application security product management at Radware, explains why conventional application protection struggles with microservices on Kubernetes and how Radware's Kubernetes WAAP is built for it. With 77 percent of organisations on microservices and 69 percent running Kubernetes in production, he argues that complexity is the vulnerability: dozens of components, third-party images and tools, CI/CD pipelines and many people with direct access widen the penetration surface, while 86 percent of traffic is east-west and largely unmonitored, letting malware move laterally during a dwell time that can last months. Perimeter defence and micro-segmentation are necessary but insufficient, and a service mesh enforces mTLS and authorisation yet cannot inspect the HTTP traffic that carries most attacks. The missing piece is layer-7 inspection at the microservice level, which must integrate natively with Kubernetes (Helm, GitOps, autoscaling) and add negligible latency. He describes Radware's low-footprint agent, learned positive and negative security policies, air-gapped operation and a Fortune 500 hybrid deployment. Q&A covers sidecar mode, VM workloads, latency, black-box applications and offline licensing; the session ends with a prize draw.