AI Insights
Showing results for 
Search instead for 
Did you mean: 

Eight ways DevOps will change in the future


By: Chris Riley


I'm no wizard, and I don't claim to know everything about DevOps. What I do know is that in the last one and a half years the definition of DevOps has changed dramatically, and there's no sign that its evolution is slowing. What will DevOps look like in the future?


DevOps, software delivery, software development, continuous delivery, continuous integration, datacenter, containers, microservicesThe problem is that there's not just one "DevOps." There are DevOps, the tactics, and DevOps, the movement. Chances are that if you've landed on a blog post discussing DevOps, it will be about the movement. However, most peer-to-peer conversations are about the tactics. This can make understanding DevOps highly role-dependent. The tactics are nearly synonymous with infrastructure scripting. This view, however, lacks the most complex elements—the people and the strategies that make the pipeline improve beyond infrastructure.


The movement looks at the organization holistically. It leads with one goal: faster, more frequent, high-quality releases. And the only way to make this happen is open communication, fluid processes, and a results-driven mentality. Call it DevOps or not—if you have this goal, the outcome will certainly look like DevOps.


DevOps tactics won't change dramatically in the coming years. But the movement will. Here are my thoughts on what's to come.


1. Disillusionment: DevOps, as defined by its innovators, has fizzled. It was limited to a handful of environments with applications conducive to continuous delivery that built DevOps from the bottom up. But this isn't every organization, and thus has limited adoption potential. DevOps will come out of its sophomore slump stronger, but it will leave some casualties behind.


2. DevOps is the means, not an end: If you believe DevOps is an end, you'll be disappointed once you reach it. If you view DevOps as a practice, however, you can adapt to change without reinventing your operation. When companies embrace modern processes in a static way, they become controlled by them. They may start out thinking they're a DevOps organization, but quickly realize they've actually lost flexibility and fallen behind. At which point, they either conclude that DevOps doesn't work, or they realize that DevOps is a journey, not a certificate you get and move on.


3. Full-stack deployments: Applications are becoming more complex: there are more components, the relationship between the front end and back end is tighter, and system-level bugs and exploits are harder to spot. If you look at an application as the entire stack, all these issues become easier to address and manage. In future applications, code won't be separated from infrastructure and system configurations. Rather, applications will be snapshots of all the above, and when you release, you'll release the entire stack.


4. No immutable infrastructure: Because we will have full-stack deployments, we won't release to the same infrastructure each time. In the modern application, machines will be swapped out on the fly many times a day. We first need to change our way of thinking about servers as static objects. They now can be replaced with an updated version on a whim. But this also requires special technology. Virtualization has existed for some time to make this possible, but it's heavy. Containers take virtualization to the next step—they're more nimble and not attached to a hypervisor. In the future, pipelines will be driven by containers.


5. The new container: Containers still lack the functionality to make them work well with application back ends, larger applications, and reliable networking and security. However, that's going to change fast. Now that container technology is an open standard, organizations can embed it in a range of products, such as release automation and data center management. Linux containers (LXC) might not even be the core. For example, Microsoft is finding ways to implement a similar technology in Windows Server. Newer versions of the container model will fill the gaps.


6. Microservices: Applications might even come to contain the whole data center. In the microservices model, an application is split from one monolithic collection of functionality into a front end that syndicates small services across a cluster of containers. Each container runs just one individual service. This builds in a lot of agility and scalability. Ancillary technologies such as software-defined data centers or data center operating systems will be seeing increased interest because of this trend.


7. The DevOps hub: The DevOps hub will contain all the tools that drive all the processes, continuous delivery and integration, and management tools such as log analysis and alerting. A one-stop shop but likely not suitable for the enterprise.


8. Goodbye, versions: In DevOps processes, everything is continuous. In this environment static elements can't survive. This includes versioning. There will still be versioning of code and scripts, but versioned releases will be a thing of the past. In the most advanced environments, versions won't even be identifiable, because only portions of the application will be released on every developer commit.


The future of DevOps might be even more unpredictable than we think—two years from now we might not even use the term "DevOps" at all. But we won't drop the end goal of higher-quality software, and the future will be defined by the steps it takes to reach it.


Read the case study, Through the DevOps looking glass: Learnings from HP's own transformation initiative,for details on how HP uses DevOps to optimize application release and delivery.



About the author

Chris RileyChris Riley


Chris Riley is a technologist who has spent 12 years helping organizations transition from traditional development practices to a modern set of culture, processes and tooling. In addition to being a research analyst, he is an O’Reilly author, regular speaker, and subject matter expert in the areas of DevOps Strategy and culture and Enterprise Content Management. Chris believes the biggest challenges faced in the tech market is not tools, but rather people and planning.

About the author

Connect with Chris:

 Follow me on Twitter @HoardingInfo

0 Kudos
About the Author