The Future of DevOps
More automation, ingrained security, shorter feedback loops with data being used to drive automation.
Join the DZone community and get the full member experience.Join For Free
To gather insights on the current and future state of how DevOps is scaling in the enterprise, we asked IT executives from 32 different companies to share their thoughts. We asked them, "What’s the future for DevOps from your point of view — where do the greatest opportunities lie?" Here's what they told us:
- Automation with more focus on CD. AI/ML is a perfect fit for DevOps culture. There is a vast amount of data to be leveraged to continuously improve the process and free resources with automation and machine learning.
- It will be around integrating the tools you have, tools available in open source, tools made available by other vendors, and the platform to move things closer to solving real business problems. You want more automation around deploying and installing software and the ability to take action. Scale your deployment automatically. Use the data you have to automate more of what is done. How to focus on the business rather than solving minute tech problems. How to focus on security and business problems.
- The greatest opportunity in DevOps is continuous, autonomous testing. Most CIO’s will tell you that test automation remains their largest bottleneck for implementing successful DevOps. With DevOps, testing must be continuous, and developers truly need to know the status of those tests as soon as possible or risk releasing a product that will harm a company’s reputation.
- Automation is where the greatest opportunities lie when it comes to DevOps. By using automation software, organizations will be able to further eliminate the manual, repetitive elements within the development process so IT teams can provide greater value to the business. With IT teams focusing on providing more value rather than slowly molding their way through the DevOps process, businesses can stand out in competitive markets.
- I think the greatest opportunity lies within security. I see some of the best practicing DevOps teams that still struggle with things such as Identity, Authorization, and Secrets Management. The future is in further educating and empowering these individuals and teams related to those parts of security as well as overarching risk and compliance to fully ensure the individual and the teams have the skills, access, ownership, and data to make the right decisions.
Feedback Loops Inform AI/ML
- The power tools for feedback loops need further democratization. Multi-billion dollar internet giants have amazing in-house tooling to roll out code in controlled ways and observe the outcome. Engineers that want to increase the velocity of value, of code that moves the needle for their business, should not have to go work at Booking.com or Netflix or Amazon to have a world-class experimentation platform at their disposal.
- It's true for DevOps, and many other categories of software, that it’s becoming a lot more about the data. Organizations that do DevOps are deploying code faster, but it puts more strain on security, auditing, how to collect data to do meaningful stuff. There is so much data now that DevOps will be dependent on AI/ML. Harvesting the data to make insights and informed business decisions.
- We are just scratching the surface of what we can do with advanced software development pipelines. Telemetry will be hooked up to everything. We'll see how an app was used and whether or not it caused a problem. We will roll out features to a small group and then everyone. If telemetry says it’s not working, we can roll it back. This is the future. Build tooling around that. You don’t have to build customer infrastructure. Make everything you build part of the testing process.
- The greatest opportunity is providing the speed of IT delivery we’ve come to expect in our consumer-driven culture. As private consumers, all parts of our personal lives are about getting superb customer service with defect-free results. Enterprise IT organizations struggle to deliver a similar experience to its internal customers and end-users. DevOps really gives enterprise IT organizations the opportunity to use speed to transform into technology organizations and to provide customers results rapidly without defect and with superb customer service.
- I look at DevOps as the operating model for IT. Moving upstream to product planning, portfolio management. Move outside to other parts of the organization like sales and marketing to move left and increase feedback loops. Take best practices from other industries to move throughout the enterprise. If you’re a constraint to the value stream you need to shift left and implement DevOps. IT is driving where the business is going.
- Better education is still needed more for everyone from the end-user, to developers, to executives. Help everyone understand what DevOps can bring to the organization and how it can bring the organization down if it is mismanaged. Developers are getting too sloppy with code and pushing out bad code quickly.
- DevOps provides better outcomes. This becomes evident in terms of developer career development, team performance, quality of services produced, and speed of return on investment to the enterprise. It is an optimization worth striving to establish within your organization. However, once established, an equal and sustained commitment is required to maintain and scale DevOps on a long-term basis. A future built of composable DevOps produced and supported services is undoubtedly a brighter future for all involved.
- The entry point is using DevOps to move faster and when things start breaking you then start looking at site reliability engineering (SRE). They are not mutually exclusive. As you are moving fast things break faster. SRE is able to handle scale.
- Change is inevitable. The DevOps approach is here to stay, and we need to expand and evolve with new technologies. The cloud domain is definitely growing at a rapid pace and DevOps engineers would have to up their game to stay relevant.
- Managing the people, process, and technology that goes into a successful DevOps implementation is the next big opportunity for companies to take their DevOps journey to the next level.
- Serverless is a big opportunity because it shrinks the size of a deployable unit even further than microservices did and provides a nice separation of concerns. Before microservices, implementing a deploy meant deploying the entire monolithic application. Therefore, so if a deploy went wrong, the entire application was affected. With microservices, individual components of an application can be deployed without touching the rest, which greatly limits the blast radius. Serverless goes one step beyond microservices and will continue to reduce the risk of failed deploys, creating fast iteration enabled by DevOps.
- DevOps is exciting because there will never be a point when there won't be a need for innovation. As people build more powerful platforms, new problem sets will arise. For example, once containerization problems solve for standardization and now that there are powerful tools aimed towards orchestration. Istio was a good example of something that comes out after orchestration; if you don't have a massive number of containers, Istio (which operates at the network level) won't be useful. I'm personally interested to see innovations occurring in integration security into DevOps, and having more source control awareness in DevOps (e.g., catching errors before they enter your infrastructure.) as we move more towards Infrastructure as Code.
- DevOps methodology will continue to be adopted by other departments beyond the software development and operations teams. Just as the agile scrum concepts have been found useful in content marketing, documentation, design, etc., so too will those teams seek to automate and orchestrate their workflows. In addition, they will seek to partner with upstream and downstream teams in the same way that DevOps brought development and operations (and more recently, security) teams together. We may have to rename DevOps at some point when so many groups outside of Dev and Ops are using it!
- DevOps toolchains will continue to become more complex. As more parts of the organization use DevOps tools for more types of workflows, as more legacy applications get onboard into DevOps systems bringing proprietary code and processes with them, and as organizations seek to orchestrate and automate broader swaths of cross-functional workflows beyond the previous scope of DevOps, the DevOps tools will reflect this complexity. More plugins, more configurations, more users and teams with more complex permissions, and more variety in development workflows and release processes will be the continuing norm. DevOps experts will be tasked to create and maintain order and will look to their tools to simplify access control management, handle broader types of applications in the same workspace, adapt more easily to legacy applications, and be flexible and extensible enough to stretch to use cases that may not have been in the original scope.
- Companies are looking to empower developers to control their own pipeline logic by giving them automatically provisioned resources in which to execute those pipelines. To industry leaders, it’s no longer tolerable to maintain stand-alone build systems that require one-off provisioning and separate security controls. As operators seek to remove themselves as bottlenecks, teams are moving towards systems they can operate centrally, but that provide distributed power to teams to self-provision and auto-scale on their own.
- Many of the thorniest problems in DevOps land are increasingly being abstracted. Common challenges like virtualization of environments, orchestration of infrastructure, and scheduling execution over complex sets of processes used to require manual intervention or were glued together with fragile scripts. Today, we have tools like Docker, Kubernetes, Terraform, and scores of others that model and reliably automate such problems. The best IT teams are always looking to shrink the surface area of esoteric risk. This is why they opt for industry-standard and/or vendor-supported abstractions for common problems over attempting to create, maintain, and innovate on bespoke systems. High-performing software organizations focus as much energy as possible on proprietary value creation rather than the supporting systems that are common problems among their peers. They favor open standards that provide expressive codifications into common syntax and tooling.
- As automation in software delivery pipelines has grown, compliance efforts have struggled to keep pace. Companies investing in software delivery throughput as a strategic asset don’t want that investment stymied by business process bottlenecks. Operators are looking to assure their various stakeholders that they have strong practices like segregation of duties, prudent security key management, high availability, disaster readiness, and the like. Systems that bake such concerns into their core operating model without compromising team throughput succeed because they protect operators’ most precious asset: the stability and accountability of their systems.
- The future is when we don’t need the term DevOps anymore. The processes and tools which we now lump together under the term DevOps are simply being employed on a day-to-day basis, without anybody making a big fuss about it anymore.
Here’s who shared their insights:
- Nancy Wang, CEO, Advancing Women In Product
- Mick Morrissey, Director of Engineering, Asavie
- Lyon Wong, Founder & COO, Blameless
- Patrick Reister, Senior Build Engineer and Ivan Szatmári, Head of Release Management, Bohemia Interactive Simulations (BISim)
- Rob Zuber, CTO, CircleCI
- Brian Dawson, Director of Product Marketing and Brian Nash, Director of Product Marketing, CloudBees
- Eric Robertson, V.P. Product Marketing Management & Strategy Execution, CollabNet VersionOne
- Jeff Williams, Co-founder & CTO, Contrast Security
- Mike Rose, VP Engineering, Cybera
- OJ Ngo, CTO, DH2i
- Chris DeRamus, Co-founder & CTO, DivvyCloud
- Tobi Knaup, Co-Founder & CTO, D2iQ
- Andi Grabner, DevOps Activist, Dynatrace
- Antony Edwards, COO, Eggplant
- Kris Lahiri, Co-Founder & VP Operations and Chief Security Officer, Egnyte
- Chris Michael, DevOps Engineer, FileCloud
- Tamas Cser, Founder & CEO, Functionize
- Justin Stone, Senior Director of Secure DevOps Platforms, Liberty Mutual
- Mark Levy, Director of Strategy, Software Delivery, Micro Focus
- Phaedra Divras, Chief Operating Officer, Mission
- Michael Morris, Senior Director of IT Cloud & DevOps Platforms, NetApp
- Tori Wieldt, Developer Advocate & Sr. Solutions Marketing Manager, New Relic
- Bob Davis, CMO, Plutora
- Veejay Jadhaw, CTO, Provenir
- Vishnu Nallani Chekravarthula, V.P. Head of Innovation, Qentelli
- Anurag Goel, Founder & CEO, Render
- Davy Hua, Director of DevOps, Security & ITOps, ShiftLeft
- Dave Karow, CD Evangelist, Split
- Ben Newton, Director of Product Marketing, Sumo Logic
- Adityashankar Kini, V.P. Engineering, Sysdig
- Neil Barton, CTO, WhereScape
- Dan Beauregard, DevOps Evangelist, Xebia Labs
Opinions expressed by DZone contributors are their own.