Category Whitepapers and Guides
Continuous Integration (CI) and Continuous Delivery (CD) have made some fundamental changes to traditional development and operations paradigms. Rather than delivering builds weekly or monthly, developers have to release updates and new versions on a daily basis – in fact, it’s common for developers today to push out multiple CI builds in a single day. Consequently, operations engineers today have to deal with significantly more environments and instances, and being able to test and validate these builds as quickly as possible is key to maximising the benefits of DevOps. As many of these new builds may include configuration changes, it’s no longer practical to implement each of these individually. To this end, infrastructure automation is an invaluable component of DevOps, and in this blog we’ll discuss what infrastructure automation entails, provide some tips and best practices for implementing infrastructure automation, and talk about the concept of infrastructure as code.
What does Infrastructure Automation entail?
To understand infrastructure automation, it’s important to first discuss two key concepts:
In his blog, Sanjeev Sharma defines cycle time as “The average time taken from the time a new requirement is approved, a change request is requested or a bug that needs to be fixed via a patch is identified, to the time it is delivered to production.” Essentially, it is the total time taken from the beginning to the end of a process, as agreed by you and your customer.
Because developers require multiple environment configurations and patch levels on-demand in a DevOps environment, operations engineers need to reassess the way they manage change and maintain their environments. Any change made to an environment, from changing the configuration to applying an update or patch, should be considered a new environment version. The only way that this can be managed in practice is through scripts, which create a new version of the environment when executed.
So what exactly is infrastructure automation?
The ideal solution to both minimising cycle time and efficiently and reliably versioning environments is to capture and manage your infrastructure as code. Processes like setting up a new environment or updating an existing environment become a simple matter of:
Infrastructure automation, in other words, allows you to expedite environment management while maintaining best practices for .
How does infrastructure automation bring efficiency to my organisation?
Kent Beck, software engineer and creator of Extreme Programming, claims not to be a great coder, but a good coder with great habits. Essentially, that’s the aim of infrastructure automation – to mitigate the chance for error and enhance the productivity of development and operations through keeping good habits. Though many organisations have adopted agile methodologies, the reality of agile adoption in many businesses today is a far cry from the original agile manifesto. Many organisations operate under the pretence of being agile, but in practice are closer to what Forrester terms ‘Water-Scrum-Fall’. The reason for this is a lack of organisation wide adoption of DevOps, with agile adoption typically being limited to the development team, while operations and production run at an entirely different speed. Without implementing infrastructure automation and approaching infrastructure as code, the true benefits of DevOps cannot be achieved.
Four ways to increase efficiency in IT operations through infrastructure automation are:
Infrastructure automation simplifies many of the tasks that operations engineers have to deal with daily while keeping best practices intact. By automating wisely, you’ll be able to increase operational efficiency by several orders of magnitude without compromising on quality.
Breaking your infrastructure down into modular components and making them available from an infrastructure automation tool makes it easy to continuously manipulate systems without affecting development – for example, duplicating the production environment for bugfixing whilst provisioning multiple instances of the production+1 environment configuration for testing.
There are several open source infrastructure automation solutions available – Puppet and Crowbar, for example – that are constantly being optimised. Opting for open source solutions is a good way of avoiding tying yourself in to software that starts out strong but ends up being stitched together by TODOs, FIXMEs and quick hacks. At the time of writing this article there have been over 73,000 commits by more than 2,000 developers in the Puppet community. Few enterprise software vendors can match this pace of development.
Keeping a meticulous record of version changes to your environments, no matter how small, is essential to avoiding potentially crippling catastrophic system failures. As Jesse Robins put it, the ultimate goal of infrastructure automation is to “Enable the reconstruction of the business from nothing but a source code repository, an application data backup, and bare metal resources”.
Speak to a DevOps consultancy that understands your needs.
There’s no universally correct way to go about automating your infrastructure – each organisation will have its own goals that it aims to achieve through adopting DevOps, and the tools and strategies used to do so will vary from one to the next. To get the most out of DevOps, speak to a consultant that understands your context. With over ten years’ experience in automation, ECS Digital provides end-to-end solutions to solve your business challenges, whatever they may be, in the best way possible. To find out more about what we offer, please don’t hesitate to contact us.