2026 · lightning
Event-Driven Autoscaling with Kubernetes and KEDA: Practical Strategies for Modern Applications
Kubernetes has become a cornerstone of modern application deployment, but many teams still rely on basic CPU and memory metrics when scaling their workloads.
This seminar provides a concise introduction to KEDA, Kubernetes Event-Driven Autoscaling, and how it expands Kubernetes scaling beyond traditional resource-based signals. Designed for audiences who may not be working with the latest cloud-native stacks, the session will focus on approachable, real-world autoscaling patterns that can improve reliability, responsiveness, and resource efficiency.
We’ll look at the kinds of workloads where event-driven scaling can be useful, including backend APIs, background workers, and integration-focused microservices that depend on systems such as Kafka, queues, databases, or streaming platforms. Rather than walking through a full implementation, this session will describe the core concepts behind KEDA, the types of external signals it can use, and the practical considerations teams should keep in mind when deciding where event-driven autoscaling fits.
By the end, participants will have a clearer understanding of how KEDA can help Kubernetes applications respond dynamically to demand, and how these patterns can support more cost-effective and adaptable service architectures.