In the ideal development practice, we secure, shape, and observe traffic between services with a single line of code. However, most environments have multiple types of applications running many versions across diverse workloads and platforms, from containers to public cloud to private datacenter. With so many platforms and application frameworks, you cannot use the same code libraries across all services to shape traffic, secure communications, or enhance observability. How can we reduce the development and operational complexity?
In this session, you dive into why and how a service mesh can alleviate the management complexity of shaping, securing, and observing traffic across multiple platforms and environments. First, I’ll provide a short introduction to the session’s setup, which uses HashiCorp Consul and Envoy proxy on Kubernetes. Then, you will learn how to implement and debug traffic shaping and certificate management in the mesh. Finally, you will configure tracing and metrics collection for your service mesh application and examine the telemetry in Prometheus and Jaeger. We’ll compare the experience of using a service mesh to various programming language implementations and discuss how to extend the mesh across different workloads.