5 min read

Pre-KubeCon EU workshops on Kubernetes dev tooling

The Upstream Azure Container Compute Team is headed to Barcelona in a few weeks for KubeCon + CloudNativeCon EU 2019! Join us the day before the conference starts for FREE open source dev tooling workshops at the Hotel Porta Fira (right by the conference venue) on Monday, May 20th.

Get hands-on learning on our open source Kubernetes tools, including Helm, Virtual Kubelet, Brigade, and CNAB, and work with Windows Containers on Kubernetes. Read below for the agenda, session descriptions, and how to sign up. We have multiple sessions running throughout the day, so you’ll need to sign up for each session individually. Stay all day or pop in for specific workshops. Space is limited so sign-up now!


Workshops Agenda

Session descriptions

Pack Your Bags: Managing Distributed Applications with CNAB

When we deploy to the cloud, most of us aren’t dealing with just a single cloud provider or even deployment tool. It seems like even the simplest of applications today need nginx, Let’s Encrypt, persistent file storage, DNS, and somewhere in there your application.

That is a lot to figure out! Porter, a cloud native package manager built on CNAB, helps you manage everything into a single package and focus on what you know best: your application. In this workshop, we’ll first cover the CNAB specification. Next, we’ll deep-dive into Porter and show how it simplifies the bundle authoring experience. Finally, we’ll walk through several hands-on examples and provide you the basics so you can explore CNAB and Porter to build bundles for your own projects.

  • Audience: Intermediate DevOps/SysAdmins
  • Pre-reqs: Familiarity with Docker. Attendees should bring a laptop with Docker installed. We’ll provide Github repos for the workshop material.
  • Instructors: Scott Coulton, Cloud Advocate
  • Time offered: 9AM – 12PM

>Sign up for this session HERE.


Getting Started with Helm, the Kubernetes Package Manager

Get started with Helm in this hands-on workshop and learn how it helps you manage applications in Kubernetes. From installing Helm to working with repositories, we’ll get your started with this standard piece of the Cloud Native ecosystem. Next, you’ll learn how to write and use Helm Charts. We’ll take an existing application and package it up as a Helm chart. Along the way, we’ll learn how to structure charts, how to write Helm templates, and how to test out your chart.

  • Audience: Beginner
  • Pre-Reqs: Basic knowledge of Kubernetes, basic scripting (any language), your laptop
  • Instructors: Taylor Thomas, Azure Software Engineer
  • Time offered: 9AM – 12PM

>Sign up for this session HERE.


Deploying Virtual Kubelet to The Cloud + Autoscaling with Virtual Node

Let’s take a peek into all the new features of Virtual Kubelet, like service discovery, GPU support and more. We will also discuss the current state of Virtual Kubelet supported providers with some design decisions that were made in the past year. Next, we’ll learn about Virtual Node, an easy and simple way to scale your services up zero and up from zero to Kubernetes. We will deep drive into the architecture of Virtual Node and figure out how to autoscale Kubernetes deployments with it.

  • Audience: Level 300, DevOps Engineers
  • Pre-reqs: A good understanding of Kubernetes and the current ways to use the horizontal pod autoscaler
  • Instructors: Ria Bhatia, Azure Program Manager
  • Time offered: 1 PM – 2:50 PM

>Sign up for this session HERE.


Windows Containers in K8s

Kubernetes 1.14 brought a highly anticipated feature – production-level support for Windows workloads. If the same tools and processes can be used for Windows and Linux workloads, then they should! This workshop will introduce Windows Server containers, and then give you hands-on experience working with them on a Kubernetes cluster. We will cover Windows container differences from Linux containers, Windows workload considerations, pod specs for Windows containers, running Windows containers as services, and troubleshooting steps.

  • Audience: Enterprise application developers and operators with Windows workloads
  • Pre-reqs: Basic knowledge of Kubernetes and Windows
  • Instructors: Craig Peters, Azure PM // Ralph Squillace, Azure PM
  • Time: 1 PM – 4 PM

>Sign up for this session HERE.


Instant scaling for your Kubernetes pipelines with Brigade and Virtual Kubelet

Brigade is a CNCF Sandbox Project that allows you to construct pipelines as a response to an event (push to a git repository, an HTTP webhook, or any custom event source).  With Brigade, you are able to share data between all steps in your workflow, and handle application errors using a real programming language – JavaScript. When an event is triggered, Brigade schedules the containers in your Kubernetes cluster.

But what happens if you have a sudden spike in events that need to be handled, since provisioning infrastructure for unpredictable loads is often inefficient, from a cost perspective – using Virtual Kubelet, we can schedule all pipelines to run in Azure Container Instances, and only pay for the time when you actually process your events.

In this workshop, we will learn how to start working with Brigade, and create our first workflow, then setup Virtual Kubelet to instantly scale our pipeline to handle all events, while also learning how to start writing distributed pipelines, and execute our processes concurrently.

Finally, we will look explore the Visual Studio Code extension for Kubernetes, Helm, Draft, Brigade, and Virtual Kubelet, and see how best use these tools to streamline your development process on Kubernetes, for both the inner-loop development process (as we work on code, but before pushing to version control), as well as the outer-loop, and integrating with various CI systems.

  • Audience: Developers who build applications on Kubernetes
  • Pre-reqs: familiarity with at least one programming language
  • Instructors: Radu Matei, Azure Software Engineer
  • Time offered: 3 PM – 5 PM

>Sign up for this session HERE.


Serverless platforms have the “magic” ability of scaling your applications from zero containers to as many as needed based on incoming events. They also scale back to zero containers when no more events are coming through. The recently announced KEDA allows you to transform your Kubernetes cluster into a true serverless platform. For you to take advantage of this event-driven scale you need to run a workload that can process events like Azure Functions.
In this workshop, we’ll cover KEDA concepts and how it works. Next, we’ll cover how to create a simple application with Functions and package it in a container. After that’s done, we’ll give you a chance to deploy an end-to-end solution that can listen to an event source, like Kafka, and process your own Functions code.
  • Audience: Developers who build applications on Kubernetes
  • Pre-reqs: Basic knowledge of Kubernetes and application development
  • Instructors: Eduardo Laureano, Azure PM // Jennifer Lee, Azure PM
  • Time offered: 4 PM – 4:50 PM

>Sign up for this session HERE.