Service mesh in production with Linkerd

Charles Pretzer (Buoyant)

9:00am–12:30pm Monday, February 24, 2020

Location: Beekman Parlor

Microservices

Secondary topics: Best Practice, Framework-focused, Hands-on

Average rating:

(5.00, 2 ratings)

Who is this presentation for?

Kubernetes application developers, site reliability engineers, engineering managers, DevOps practitioners, and release managers

Level

Intermediate

Description

In the technology landscape, cloud innovation has been a substantial driver in the migration to distributed systems. Application development has followed suit, and monolithic applications are being broken down into smaller, more manageable pieces. These smaller pieces require the same level of monitoring, security, and management as their larger counterparts, and as the number of services grows, so does the complexity of managing the health of the overall application.

A service mesh adds a layer on top of the orchestration infrastructure that’s used to manage the microservice application and simplifies application management while conserving and encapsulating complexity. Charles Pretzer provides the knowledge and tools to enable you to successfully implement a service mesh using Linkerd on a demonstration application. You’ll leave with a basic understanding of how a service mesh works, how to deploy Linkerd to your own Kubernetes clusters, and how to configure Linkerd for ongoing application maintenance and operation.

Outline:

Review of microservice architectures and service mesh concepts

Lecture: The past and present of distributed systems and how they apply to microservice architectures
Hands-on exercise: Install the demonstration application on a Kubernetes cluster

Introduction to Linkerd—Design and concepts

Lecture: Observability, security via mutual transport layer security (TLS), automatic sidecar proxy injection, and more
Hands-on exercises: Deploy Linkerd to the cluster, collect metrics via the dashboard and CLI, deploy service profiles to the application for fine-grained metrics and traffic control, troubleshoot errors in the service, and troubleshoot the Linkerd service mesh

Day one service mesh with Linkerd

Lecture: Continuous integration and continuous delivery, deployment methodologies, canary, blue-green, the service mesh interface, and Flagger
Hands-on exercises: Implement traffic split to route traffic to different versions of a service based on weight; read distributed tracing data from the Linkerd proxies into a UI

Prerequisite knowledge

Familiarity with microservices architectures
General knowledge of networking concepts, especially in Kubernetes
Experience using kubectl to deploy applications to Kubernetes clusters

Materials or downloads needed in advance

A WiFi-enabled laptop with kubectl and Git installed and access to a Kubernetes cluster, including Minikube or microK8s
A GitHub account

What you'll learn

Explore service mesh and how to implement Linkerd on your Kubernetes clusters
Understand the need for observability, security, and control in a real-world application
Learn what you need to take back to your managers and colleagues to explain the benefits and trade-offs of using a service mesh

Charles Pretzer

Buoyant

Charles Pretzer is a field engineer at Buoyant, where he spends his time collaborating and engaging with the open source community of the CNCF service mesh Linkerd. He also enables production-level adoption by helping companies integrate Linkerd into their Kubernetes-based applications. Charles has spoken at meetups and conferences hosted by ABN AMRO, Macnica, and NGINX Conf. When he’s not presenting or in front of a computer, he’s riding a motorcycle or making a delicious mess in the kitchen.