Formed in 2009, the Archive Team (not to be confused with the archive.org Archive-It Team) is a rogue archivist collective dedicated to saving copies of rapidly dying or deleted websites for the sake of history and digital heritage. The group is 100% composed of volunteers and interested parties, and has expanded into a large amount of related projects for saving online and digital history.
Infrastructure-as-Code (IaC) practices, which in turn are made possible through infrastructure or platform clouds (IaaS or PaaS), solve many important issues related to continuous delivery and DevOps. Therefore, the implementation of IaaS and PaaS in a private, hybrid or public cloud model should always be aligned with your DevOps initiative. In a previous article, I argued that OpenStack is the ideal habitat for DevOps. After all, OpenStack is backed by some of the biggest companies in software development and hosting, as well as thousands of individual community members. Indeed, many argue that OpenStack is the real future of cloud computing. However, that cloud environment might also be based on proprietary VMware technology, hybrid Amazon Web Services (AWS) or similar. Whichever you end up choosing, the cloud is an important enabler for companies looking for continuous delivery in agile environments.
Why? Why is IaaS an ingredientfor a successful DevOps strategy? Through IaaS and ultimately PaaS, the bottleneck of infrastructure provisioning gets resolved; No more dependency on busy ops teams with long task backlogs and not enough of people, who are able to script and automate. Infrastructure provisioning is required to create all sorts of environments on-demand, including development, test and integration environments, all of which are required in different phases or steps within the continuous delivery chain or release pipeline. These environments can be created in isolation and dependencies from other projects, which are responsible for so many delays in a release, disappear.
Test automation is critical for continuous delivery, but benefits are diminished if the provisioning of the necessary test environment takes days or weeks.
Tips for DevOps-cloud alignment
So we’ve made the case for the cloud as part of a DevOps strategy. How do you now go about aligning your DevOps and cloud initiatives together? What factors do you need to take into consideration as your developers instantly spin up environments at any time?
First, start with IaaS. IaaS is easier to implement and more commonly applicable than PaaS. For quick wins, use public IaaS such as through AWS or Microsoft Azure. Get the necessary compliance and security aspects for dev or test environments settled and create a couple of deployment scripts on top of these services. Either prepare ready-backed images with middleware, application servers and databases or leave it to developers to install these components on their own. Alternatively ops can also prepare necessary install scripts and provide them to development.
Second, build your own IaaS, based on OpenStack or other technologies. The toolsets and interfaces used by the agile development teams and the team responsible for the IaaS infrastructure need to be synchronized. For example, if you’re using OpenStack for the cloud environment, align your provisioning tools around that, possibly using Chef or Puppet. The good news is that many of these technologies really work together well.
Third, once your IaaS implementation is mature and has been adopted, move towards PaaS. PaaS is the champion’s league in private cloud and usually requires an organization to commit to certain standards for databases, middleware, application servers, deployment architecture and more. It makes it super-easy and fast to provision not only a server, but really the complete runtime environment of an application and deploy the application on top of it.
In all phases there needs to be tight collaboration between the departments. Your agile development, IT operations and infrastructure teams responsible for delivering the private cloud all need to be aligned around the common goal: delivering quality apps and updates to market more quickly. That means working as one to build a reliable, fast and adaptive cloud infrastructure in support of agile services delivery. It is imperative to understand and track individual development milestones. There’s no point in relying on the ‘big bang’ approach: the cloud infrastructure needs to be synchronized with all the development timelines.
It has been proven particularly useful to use agile methods in structuring the infrastructure teams: Each platform gets a product owner, who works with development teams and manages all the requests developers have for the platform, i.e. the required APIs and their signatures or supported versions of specific technologies in a common backlog. This is done in the same way as product owners in development are managing their backlogs for the business.,
Finally, make sure you have the right skills in place. DevOps and private cloud infrastructures command new skills sets—operations engineers, for instance—to prepare, manage and develop your private cloud and the application platform. There’s no point in buying in skills half way through the agile development process: they need to be built into the business plan from the start.
DevOps,IaaS and PaaS go together like the PC and the keyboard; apps and the iPhone; Windows and Word. Together, DevOps within an IaaS infrastructure enables you to harness faster development and move to continuous delivery.
