In the past several years, Kubernetes has become the de-facto standard for orchestrating containerized, stateless applications. Tools such as StatefulSets and Persistent Volumes have helped developers build stateful applications on Kubernetes, but this can quickly become difficult to manage as an application scales. Tasks such as machine learning, distributed AI, and big data analytics often require a distributed application to maintain some sort of state across services. 

KubeDirector is an open-source controller that helps streamline the deployment and management of complex stateful scale-out application clusters on Kubernetes. KubeDirector provides an application-agnostic deployment pattern and enables developers to run non-cloud native stateful applications on Kubernetes without modifying the code. KubeDirector aims to bring enterprise-level capabilities for distributed stateful applications to Kubernetes.

Kartik Mathur is an engineer at HPE Developer, an open-source initiative within Hewlett-Packard Enterprise. HPE is an enterprise contributor to the KubeDirector open-source community. Kartik previously worked as senior software engineers at BlueData, which created the KubeDirector project before its acquisition by HPE. Kartik joins the show today to talk about why state is important for Big Data or Machine learning applications, how KubeDirector can help manage the complexity of stateful applications.

Sponsorship inquiries: [email protected]

The post KubeDirector with HPE’s Kartik Mathur appeared first on Software Engineering Daily.