"This was my first LISA. It was great attending a conference that focused on my role without trying to shoehorn every challenge into a specific vendor's solution. The same mix of awesome sessions, speakers, and other attendees will bring me back every year."
LISA14 Attendee
"Great sessions, great presenters, great community. I feel like I can make up a year of following news and forums on the subject in a week."
LISA14 Attendee
"I like seeing where the industry is heading, how SysAdmins/engineers are evolving, tools they are using, common problems and solutions across the world."
LISA14 Attendee
"LISA is the best mix of training, talks, and networking of any events I've been to. That's the reason I've been to 11 of them.”
LISA14 Attendee
"In a world where technology changes rapidly, it's hard to find great resources for in-depth expertise. LISA has a terrific combination of people who create the technology and people who implement it effectively for their companies."
Matt Simmons, Northeastern University
"If you're a sysadmin slaving away in a metaphorical basement and re-inventing the wheel every time your employer's business goals change slightly, LISA will help pull you into the light."
Marc Chiarini, Long-Time IT Admin and Researcher, MarkLogic Corp
"No matter if you are dealing with the latest tech or trying to maintain something from the dark ages of the ‘90s, there are people at LISA who are experienced with it. Heck, many of the people who _developed_ this tech attend. This is a great place to get answers and ideas."
Lee Damon, University of Washington
"LISA is the home of ops people who do ops correctly. I’m always motivated by the conference content and hallway conversations to be better, and I take home the knowledge needed to move closer to that goal. "
You have installed Docker, you know how to run containers, and have written Dockerfiles to build container images for your applications (or parts of your applications). Now comes the next part: connecting multiple containers together and deploying them on a cluster of Docker hosts instead of your single development machine.
This is what we will cover, along with some production-related questions: How to manage logs? Backups? Remote access? Security upgrades?
In this workshop, we will present the networking model of containers and cover how to run containerized applications spanning multiple Docker hosts. We will give recipes to implement cleanly various ops tasks including logging, backups, and more. Finally, we will discuss orchestration and scheduling. We will present solutions like Swarm and Mesos, and see some of them in action on an actual microservices architecture.
Who should attend:
Developers and system administrators who are familiar with basic Docker concepts and want to learn how to deploy Docker for real-world applications.
Take back to work:
You will know how to deploy Docker "for real," and how to perform common operational tasks with containers. You will learn about patterns and best practices to scale applications with Docker and containers.
Topics include:
Pre-requirements
VM environment
Our sample application
Running the whole app on a single node
Finding bottlenecks
Scaling workers on a single node
Scaling HTTP on a single node
Connecting to containers on other hosts
Abstracting connection details
Backups
Logs
Security upgrades
Network traffic analysis
Introducing Swarm
Setting up our Swarm cluster
Running on Swarm
Network plumbing on Swarm
Last words
Additional Materials:
You won't need to pre-install Docker before the course; each student will be given credentials to connect to an individual virtual machine. All you need is a computer with an SSH client.
