3 minute read

Lots of containers this week, with content and discussions from DockerCon and ContainerSched front of mind. Other content covers some ideas about meetings and deployment cadence, and several interesting new software projects.


Devops Weekly is sponsored by Brightbox Cloud: the multi-zone cloud platform built for High Availability.

Try an SSD cloud server in 30 seconds with your £20 free credit…


[Webinar] How to Build a Metrics-optimized Software Delivery Pipeline

Learn a practical step-by-step approach to optimizing your delivery process so you can deploy better quality software faster! You’ll learn why you should move to a metric-driven pipeline, which key quality metrics to measure and how to integrate them to catch problems earlier.


Interesting observations about the relationship between meetings and shipping code. Some other good examples of the impact of the size of deployment on frequency of deployment too. I’d love to see that studied with data from the field.

One of the closing demos from DockerCon EU was this example of running unikernel applications (in this case based on the Rump Kernel project) inside Docker containers. The post covers the why and how of the demo.

I spend the start of the week at DockerCon EU. This write-up covers some of the high level themes, including the improving Docker security story and the renewed interest in Swarm.

I occasionally include academic papers in the newsletter, and this presentation is a great reason why. The talk skims through 30+ years of research into distributed systems, picking out key concepts and papers for folks to read.

A good set of videos from the recent Security at Scale conference, covering everything from data visualisation, mobile malware, crypto, open source and more.

A presentation entitled The Road To Unikernels, from the recent ContainerSched conference in London. A good introduction to why unikernels are interesting and a look at the current state of development.

A nice write up of one organisation’s move to use microservices, some nice ideas about resource discovery vs service discovery and the advantages of a shared language for integration.

A good series of guest blog posts looking at some aspects of configuration management in a containerised world. Most of this still talks about build time management, but the series is still looking for more posts for the next week.

A detailed description of one organisation’s development environment, including details of Docker, Vagrant, Terraforma and CircleCI.

A good look at (mainly business) trends around cloud computing, looking at everything from enterprise software companies, container adoption, company spending and future trends.


Spinnaker is an open source, multi-cloud continuous delivery platform. It provides a nice looking dashboard based around pipelines, specifically designed for deploying infrastructure in cloud platforms. The video provides a great high-level overview.

Firmament is a new cluster resource scheduler intended to make high-quality scheduling decisions. It’s based on original research into flow networks and is guaranteed to find the optimal assignment based on user provided policy.

Containerbuddy provides a very nice way of integrating existing applications with containers and service discovery. It acts as PID1 inside the container, and integrates with Consul using a simple configuration file.

Heroic is a new scalable time series database based on Cassandra and Elasticsearch. It has some nice features around a global federation, allowing independent clusters to query each other.

AppArmor is a powerful security control, but configuring it can be a dark art. Bane therefore looks a useful tool, providing a much more human friendly to AppArmor configurations, with some nifty docker security profile integration.

Devops Weekly is sponsored by Brightbox Cloud: the multi-zone cloud platform built for High Availability.

Try an SSD cloud server in 30 seconds with your £20 free credit…