Choose your fault domains carefully: Mesh app architectures for K8s

TALK ABSTRACT:
Cloud Native infrastructure opens the door for extremely flexible application architectures. However, this flexibility isn't free - architects need to consider what happens when various components are unavailable. This unavailability could be due to failure, network partitioning, lack of capacity, upgrades or other factors.

We'll use the concept of a fault domain to partition availability across different levels, from "containers in a cluster" to "data centers distributed across the planet." This helps organize our approach to leveraging technologies like Kubernetes and service mesh to address availability concerns.

SPEAKER BIO:
Andrew Jenkins, CTO | Aspen Mesh (F5 Incubation)

As CTO and co-founder of Aspen Mesh, Andrew advocates and builds efficient, flexible and cloud native application architectures. You can find him presenting and listening at meetups, webinars, conferences and contributing podcasts, interviews, and articles or standing in front of a whiteboard with anyone interested in solving hard distributed problems. He's done research for NASA, wrote code for a successful startup and provided technical leadership and architecture for teams in large organizations. You can find software and hardware that he's built powering a public cloud or data center near you, or out beyond Pluto measuring the solar system.