How enterprise DevOps helps build new strategies on old foundations

ecs-admin 11th April 2016

Being able to deliver software at a higher quality and more frequently is becoming increasingly critical for IT enterprises in today’s hyper-competitive markets, and DevOps makes it possible to achieve this in many ways. But the principles that underpin Agile and DevOps aren’t only relevant for optimising software delivery and deployment chains: because enterprise DevOps relies significantly on automating complex and cumbersome manual tasks, it also provides a means of reducing the time and cost requirements of legacy technologies and systems.

There are two schools of thought on this topic. On one side are those subscribing to what Gartner call “Bimodal IT”. On the other side are those of us that believe and increasingly have the data to back up our beliefs that high performing organisations are those that adopt the practices of Agile, Continuous Delivery and DevOps across all systems.  Jez Humble recently published a good articleon the flaw in bimodal IT. In this blog, we’ll look at how and why enterprise DevOps makes it easy to build a modern IT strategy using old foundations.

DevOps isn’t a software tool or a technology – it’s a cultural shift

There are many misconceptions about DevOps, exacerbated, no doubt, by the fact that it is notoriously difficult to define and can apply to a wide range of practices. Broadly speaking, DevOps is the combination of people, processes and tooling used in the delivery and maintenance of software products, that in concert result in faster, more efficient delivery of better software. This means that ‘going DevOps’ doesn’t necessarily require a massive investment in new IT infrastructure – within the bounds of reason, it’s possible to implement DevOps in older systems equally effectively.

Change “because this is the way we do it” to “we have to improve this”

We frequently have conversations with large enterprises about their current processes and discover many challenging practices which we begin to break down by simply asking the question “why?”. Large enterprises are not new kids on the block, they have been around for a long time and providing a service long before many of the household name unicorns! They have seen technical debt build up over time which often gets brushed under the carpet. This closely relates to the psychology of the 5 monkey experiment. Human behavior means we may not question something because everyone else is doing it that way or it’s always been done that way. However with DevOps we have to begin to challenge some of these processes and practices to enable greater automation and efficiency.

Supplementing legacy systems with DevOps practices gives them a new lease on life

When it comes to Continuous Delivery and Continuous Integration, the only limits are your willingness to go head-to-head with your competition in terms of speed, quality and consistency – there’s no need to throw out ageing hardware (which is helping keep the lights on) when you can outsource part of their workload to cloud-based and as-a-Service applications. With Infrastructure as Code and Immutable Infrastructure a growing trend among enterprise DevOps implementations, it’s possible to build faster environments by physically adding machines (horizontal scaling) and applying automation through full stack provisioning allows repeatable and dynamic creation of environments on demand. What’s more, versioning every deployment individually means that rolling back to a stable version is easier than ever.

Mainframe administration can benefit from automation too

Finding new administrators for legacy mainframe systems is an uphill battle for many enterprises today – while senior admins with invaluable management skills are becoming increasingly hard to find, and both budgetary and logistical requirements make training new staff problematic. As such, admins with experience in z/OS Environments as opposed to Linux and Windows can be incredibly hard to find. But for forward-thinking IT enterprises, this shouldn’t pose a serious problem – as Wayne Kernoshan says in his blog on TechTarget, “The problem isn’t that new hires aren’t familiar with the complex, custom daily tasks of mainframe ops. The problem is that you’re still doing them.” Rather than spending time and money recruiting and training up new hires on obsolete systems, enterprise DevOps makes it possible to automate frequent tasks that require specialist knowledge to complete. And, because DevOps relies on Continuous Integration and Delivery, making smaller, incremental changes makes IT staff’s jobs easier and results in greater productivity and efficiency for both developers and admins.

Forest is a DevOps consultancy with over 12 years’ helping enterprises around the world deliver software faster, better and at a lower cost through the adoption of DevOps and Continuous Delivery practices. Our team is made up of experts from diverse backgrounds, and we can help you transition to a DevOps culture in your organisation and ensure you get the associated benefits. To find out more, please feel free to contact us.

Image Credit:

Found this interesting? Why not share it: