DevOps Trends
DevOps has been around for many years now and is typically well understood as breaking down the silo between development and operations to achieve business agility through increased velocity and building better digital reputations through increased quality.
As we continue to address the flattening of silos through people, processes, and tools, there are some trends and focus areas that have been emerging. Some are a new twist on a heritage problem, some are a reexamination and closer inspection of a known challenge, and some are trends based on emerging technology. We want to hear your opinion on the DevOps topics trending: Cloud Native, Continuous Improvement, Security, AI, GitOps, and Design Thinking.
Do any of these resonate with you?
DevOps for Cloud Native Apps
The monolithic applications of yesteryear have been modernized, in situ, with automation to accelerate build, deployment and test. Now that cloud and control planes like Kubernetes have matured, we can use emerging technologies like Knative and Tekton to address automation in the cloud native way. Technologies are being pulled into the platform, providing a lower center of gravity for commoditized services. These commodity services are taking on core functions like service discovery, service routing, domain name services, and even services to standardize how code is packaged into immutable images and deployed into the cloud. We are being liberated from these core services so development can focus on innovative things that matter to the business.
Are you aware of these newer technologies and how they fundamentally change the mechanics of how cloud native apps are delivered?
Continuous Improvement of Application Delivery
A longstanding tenant of DevOps is to continuously improve delivery (velocity and quality) over time with incremental successes. Tool automation has been a huge accelerator to increasing velocity and we have been able to measure things like deployment cycles and quantify time saved on automated deployments. We have even been able to remove impedance in delivery through process optimization. With all the emerging tools and technology, we still see clients struggle with things like *effective* and optimized product management, timely root cause analysis of production defects, and attributing toolchain bottlenecks with inefficiencies with modern development behaviors. An example? High build failure rates or falling test code coverage could indicate an opportunity to tighten up your test driven development practices.
Are you instrumenting your CI/CD toolchain with data collectors and using data-driven decisions to continuously improve your application delivery?
DevSecOps
Not that we have been ignoring security since we have been working in this space…but the term “DevSecOps” really speaks to having a maniacal focus on security in software delivery. This is an ever increasing area of focus given our march to cloud and new architectures with microservices. We believe DevSecOps is more than security tools piped into the integrated toolchain. We believe DevSecOps starts with a “secure by design” approach to designing software addressing fault tolerance, robust coding practices and embracing ideologies like the Rugged Manifesto. Yes, we need tools to lint, perform static and dynamic analysis of workloads too. But have you considered observability and traceability in the pipeline so you can track a feature from ideation to production deployment and management? These are some of the attributes auditors like to press on. How do you ensure that the code deployed in production is indeed the same code you built in your container? Can you *prove* it?
You can learn about DevSecOps in the blog series from Andrea C. Crawford:
Is DevSecOps becoming more important to you and your enterprise?
AIOps
If we were to combine machine learning and predictive insights into software delivery, we could take velocity and quality to an entirely new level. Artificial intelligence is based on the the collection, organization, and analysis of meaningful data to learn and predict. Instrumentation of the toolchain is one channel that is essential, but what about having awareness of the operational environments, the tool infrastructure, even business and regulatory conditions that might influence your delivery. What if we could detect that an application was handling PI data and were able to route deployment of that workload to a cloud environment, accordingly? What if one public cloud provider was experiencing issues, and we could deploy to other clouds dynamically?
Are you actively looking for ways to insert AI into the delivery process?
BizDevOps
You have done a great job applying DevOps practices to streamline your team. Blockers are removed, the pipeline is humming, you’re continuously delivering faster than ever before, but …
- Are you delivering the right outcomes at the speed the market demands?
Your first response might be, “Yes, we are bringing ideas to production with speed & minimal overhead, all while meeting our quality & SLA targets.” That’s not the whole answer though — Business, Design, and Technology must come together in order to be successful. First, you must Discover the Business Opportunities and align to market demands. Envision how these align to your expertise & incorporate Design Thinking into your DevOps practices. A key to success is learning from your users as they interact with your application. Features are experiments that must be proven successful using metrics.
BizDevOps requires that success is not just about dates & SLA’s — but also about Business and Design.
GitOps
Just when you thought there were not enough permutations of DevOps…along comes a story of how DevOps principles can apply to automating the provisioning of infrastructure and configuration of platforms that applications run on. Born as infrastructure-as-code, the concept of GitOps is all about programmatic preparation of platforms and environments that are ready to receive the applications. The “Git” reference in GitOps emphasizes that even infra-as-code should be considered a value asset that should be managed just as application code should. Traditional infrastructure engineers write code, even more so, full stack engineers on their infrastructure and platform oriented activities.
Are those in your IT organization using automation and configuration management and applying it as GitOps to manage infrastructure engineering activities?
What do you think?
We want to know which topics you are most interested to integrate with DevOps: Cloud Native, Continuous Improvement, Security, AI, GitOps, and Design Thinking. Have a favorite that’s not in this list — tell us about it!
Bring your plan to the IBM Garage.
Are you ready to learn more about working with the IBM Garage? We’re here to help. Contact us today to schedule time to speak with a Garage expert about your next big idea. Learn about our IBM Garage Method, the design, development and startup communities we work in, and the deep expertise and capabilities we bring to the table.
Schedule a no-charge visit with the IBM Garage.
