4 minute read

Pretty good mix of people (why silos emerge, operations debt, the importance of context) and technical (measuring worst case performance, proxies, lots of monitoring videos) posts this week I think. Lots for everyone to read until next week.


It’s time to fix your incident management. Built for DevOps, VictorOps helps you respond to incidents faster and more effectively. Try it out for free.


Practical Tips to Start Your DevOps Journey

In this webinar replay, learn practical tips that all IT groups from Dev to Ops can use to start their DevOps journey quickly. With experience from hundreds of DevOps deployments, Andreas Grabner, Chief DevOps Activist at Dynatrace, will provide insights it would take your team months or years to learn firsthand.


An interesting take on the problem of dependency management. Looking into scaling systems via libraries or services, and the different tradeoffs involved in each.

A solid argument that, although software development and systems administration are different skills sets, some experience of both helps you be better at either because you can better understand the entire context.

A look at the infrastructure behind Twitter today, exploring the network infrastructure, how the infrastructure is divided up, the various storage technologies and some of the tooling used to run it. Always great to see real-world numbers.

Devops conversations often talk about breaking down silos, but why do they exist? ANd what organisational and team structures work in doing so in the long term? Some useful comments and good references for further reading in this post.

I’ve been a big fan of proxies since I started doing operations work. This post does a great job of explaining why having a central point of control is helpful, in particular when changing underlying architecture.

All the videos from GrafanaCon 2016 have been published. Some great content on Grafana and Graphite in particular, and open source monitoring in general.

When debugging code or infrastructure it’s important to know what to measure. This post provides a good example why percentiles can be misleading, and why measuring worst case performance is important.

A nice quick overview for anyone wanting to better understand Unikernels, addressing several high-level questions people may have on hearing about them.

Devops is still a new topic for lots of people, and this introduction is a nice place to start. It provides a bit of the history, along with explaining Devops with a focus on culture, automation, measurement and sharing.

An interesting look at continuous delivery at scale, in this case behind the search engine Bing, including lots of numbers which is always a good thing.

Some good observations about why operatings debt tends to spiral out of control, and why some of the tips from the SRE book address. I love the description of the unbounded software variation problem.

Conferences are a regular occurrence in the devops space and can be great learning experiences. This post has some tips for getting the most out of tech conferences you might attend.

A nice run through of some of the tools for deploying a Kubernetes cluster. I would have liked to know more about day-2 operations as well but still a useful source.


What is cloud native and why does it exist?

Join us February 23rd at 10:00 AM PT to find out what Cloud Native Computing is and why it matters. Presented by, Alexis Richardson, founder and CEO of Weaveworks and chairperson of the CNCF technical oversight committee.

Sign up now: https://goo.gl/AfqYjo


Find a DevOps Job You Love This Year. Try Hired and Get Access to over 6,000 Innovative Companies

Secret Escapes are looking for a DevOps Engineer to join the platform team in their Central London office, Holborn. You will be working closely with the development teams to scale the testing and deployment pipeline, automating new and existing processes and driving technical changes in infrastructure and system architecture. This is an opportunity to shape Secret Escapes adoption of DevOps.



The Cloud Native Computing Foundation will visit Portland, Seattle and Vancouver from February 7-9, 2017 to offer Kubernetes training for end users, developers, students and other community members. Presented by Dan Kohn, Executive Director, CNCF, Brian Gracely, Director of Product Strategy, Red Hat and Isaac Arias, Technology Executive, Apprenda.

OSDC 2017 (Open Source Data Center) is coming up in Berlin on May 16th to 18th. The conference will include hands on workshops as well as presentations, focusing on the technical end of the spectrum. The CFP is open until January 31st too.


Cernan is a new telemetry and logging aggregation server. It exposes multiple interfaces for ingestion (eg. statsd, graphite, line-oriented log files ) and can emit to multiple aggregation sources while doing in-flight manipulation of data.

Hellogopher is a Makefile that makes your conventional Go project build from anywhere, for anyone, with just make. Aimed more at the occasional Go programmer, it makes the point that Go developers should know and use GOPATH but it shouldn’t be the first thing they are exposed to.

It’s time to fix your incident management. Built for DevOps, VictorOps helps you respond to incidents faster and more effectively. Try it out for free.