Let op: Deze pagina komt uit ons archief.
Ga voor actuele informatie en nieuws naar onze vernieuwde website.
 

A tour of Docker - The Application Container Engine

Speaker: Frank Scholten

Abstract

Docker is a popular Open Source tool that allows you to ship entire applications as containers to a server, VM, or cloud.

Traditionally, developers write code using their own tools and hand off a software package to operations engineers, who then use their own tools to configure, deploy and manage these packages. This creates numerous problems, including:

  • Configuration drift, whereby machines have slightly different configurations and are thus expensive to maintain.
  • Different toolsets in operations and development, with tools like Jenkins and Nexus on the dev side and tools like package managers, configuration management on the ops side.
  • Business risks as the diversity of tooling and the drift of configuration seriously affect the rollbackability of applications, thus creating a culture of batch and queue and risk aversion.

Docker however, allows images to be built that are used by both dev and ops engineers all the way throughout production. Both groups can now use the same artifact, the container image, to coordinate the shipping of software, which speeds up the entire process, improves quality and reduces risks by addressing the rollbackability problem.

In this talk I will briefly explain how Docker is used to solve numerous problems before taking a technical tour of the tool. The technical tour will include showing example commands, explaining its components like libcontainer and libswarm, and a very brief introduction to Golang, the language Docker is implemented. Finally, we'll look at current trends and what's next for both Docker and the companies using it.

Biography

Frank is a lean infrastructure engineer at Container Solutions. We specialize in improving software, infrastructure and management using Docker and related tools. I like to automate things and build tools that do the work for me. I am also interested in machine learning topics and have contributed to the Apache Mahout and Apache Whirr projects. In the past I have presented at Berlin Buzzwords, FOSDEM, NLJUG and NLUUG.

Najaar 2014

2023-05-27
 
Vereniging NLUUG
info@nluug.nl
           postbus 8189
6710 AD Ede