DevOps and Infrastructure as Code
With the onset of the cultural and professional movement into today’s industry I recognize that culture and people are at the forefront and also acknowledge that automation will be critical to our success as we move to improve productivity and quality throughout the lifecycle between Development and Operations. (DevOps).
Some believe that DevOps is all about pushing development paradigms into operations others believe that it is taking the accountability for risk mitigation and production that has been traditionally owned by operations into development. It is in fact both and understanding terms such and Infrastructure as code, unit testing, integration testing, and more importantly how the integrated roles between Dev, Ops and third party vendors fit into that will help to ensure the scalability, speed and productivity that we all need to deliver quality service.
Infrastructure as code
The essence of Infrastructure as Code is to treat the configuration of systems the same way that software source code is treated. Collaboration between coders, engineers, and operational staff is required to ensure that the code and production are viewed as tied, tested and validated in forms of integrated code and Infrastructure for production. The old way of moving from one stage to another will not work in today’s environment. The code, the infrastructure and the testing has to start and end with the same people via collaborative methods. We must learn to test the same way in Dev that we do in Ops and do those simultaneously when possible. A fully integrated approach that leverages from Agile, Lean and ITSM Best Practices from start to production and into operation.
The term Infrastructure as code is not new. A decade ago we used this term but the main difference today is that then we were writing scripts and code we got better at it and we have gotten better at version control. Another big difference is that technology is more complex. Now the demand is increasing making the need for version control on the infrastructure as well as the code ever more important but important to do it fast. Things have to scale across the tools and across the infrastructure.
Code is reusable but the configuration and recipes for pairing are unique depending on business outcome requirements and the infrastructure that it is running on. As barriers of entry for reusable code are made more consistent and repeatable the more efficient we as and industry will become. We will in fact shrink the amount of customization requirements. Certainly there are a large percentage of infrastructure requirements that are standardized and that can be leveraged as the norm. We need to code to insert into those norms.
There are many vendors in the industry such a Puppet and Chef that will help to optimize code and infrastructure via automation. Automation tools are available from a wide variety of vendors for unit testing, integration testing, continual deployment and continual delivery. And…yes! They are different! The key to success is not just in the tool but understanding these concepts, the people involved and when and where to leverage those tools. For more information on these terms, or for DevOps education and DevOps Foundation Certification look at what is available here: