A Practical Introduction to Service Mesh with Linkerd

Cloud native technologies have made it easier than ever to build applications using a distributed architecture. Whether building a new application from scratch or migrating a monolithic application it is important to have insight into the health of each of the components of the distributed application, as well as control over the traffic between the components.

The term "Service Mesh" encapsulates the concepts which allow for insight into and control of a distributed system: Observability, Reliability, Security, and Traffic Management.

Observability:  Service-level golden metrics: success rates, latencies, throughput. Service topologies.

Reliability:  Retries, timeouts, load balancing, circuit breaking

Security: Transparent mTLS, cert management and rotation, policy

Traffic Management:f Canary and Blue/Green Deployments, Weighted Traffic  Distribution

Using Linkerd as an example, this talk will make the details about each of these concepts concrete. Attendees will also learn how the concepts work together to make a distributed application more stable, reliable, and robust. Once these concepts are clear, attendees will have all the tools they need to make a decision about whether a service mesh is right for their application.