10 DevOps trends to watch in 2020
Where is DevOps headed this year? Let’s examine key trends around topics including training, metrics, and tools
In 2019, we saw numerous successful DevOps initiatives across large and small enterprises. If you were involved in one, congratulations! But what can we expect for the future of DevOps in 2020?
As thought leaders in the DevOps community, we believe the following ten trends will shape the next year of DevOps across the globe.
1. Agile and DevOps will increase collaboration between technology and business functions
Agile and DevOps are grassroots movements that started within technology. In many cases, however, Agile and DevOps have not been able to break out of technology. On the other hand, Agile has been adopted in other functions, including finance, human resources, procurement, and marketing. Some senior leaders are increasingly inviting their whole organizations to “become agile.”
However, this doesn’t seem to have helped the technology community join forces with their colleagues in these other functions. Because of the competitive pressure digital is putting on organizations, in 2020 we will start to see more collaboration across functions – with agile as the conversation starter.
Encourage your teams to talk to people from different functions about their experience using agile methods.
To speed this up, encourage your teams to talk to people from different functions about their experience using agile methods. Some helpful questions include: How are you doing agile? What are you doing? What is changing for you? What issues are you having? How could we work together to help to address some of these issues? These questions will help people from different functions get to know each other as people, and collaboration will improve.
[ Are you fighting skeptics? Read also: DevOps for doubters: How to deal with 9 kinds of people who push back. ]
2. Training and improving DevOps skills will become an organizational priority
DevOps requires trying out new technologies. Recent research from the DevOps Institute found that 55 percent of survey respondents prefer to hire into their DevOps teams from within their organization. Unfortunately, many companies don’t have the necessary skills to do this, and hiring new people might not be possible due to budget restraints.
One approach is to create internal training universities. This is what the courier delivery services firm FedEx did: The company knew it did not have adequate skills in its talent pool of engineers, and this led its CIO to initiate the FedEx Cloud Dojo, which teaches its own engineers modern software development and technologies and functions as a university for FedEx. The university has reskilled more than 2,500 software programmers.
Organizations that want to use DevOps to help advance their digital transformation must make drastic improvements in training, learning, and improving skills that are essential to DevOps. We expect to see a more proactive pursuit of this in 2020.
3. Upskilling and cross-skilling will lead to the rise of the T-shaped professional
Recognizing the strained talent market, organizations and individuals will invest heavily in upskilling and cross-skilling in order to meet accelerating demands for new skills. While all IT professionals will need to become more cross-domain competent, developers, in particular, will need to add new breadth to their skills portfolio in areas such as testing, containerization, infrastructure, AI, and security.
Developers in particular will need to add new breadth to their skills portfolio.
There will also be a stronger emphasis on core (soft) skills such as empathy, customer experience, and collaboration. Silos are starting to come down in many areas, and the need for everyone to become T-shaped, with depth and breadth of knowledge, will become necessary to enable and support innovation. All of this training and new collaboration (see Trend 1, above) will lead to more workers developing new technical and professional skills and personal qualities, adding new depth and capabilities to the individuals on your teams.
4. More teams will shift mindsets from “job done” to “value realized”
Value Stream Mapping can help change the way your teams think about the definition of done (DoD) from being “I did my job” to “the value is realized.” It is one of the most effective ways of changing behaviors and getting your teams to think about the end-to-end lifecycle of what they’re working on.
This is why the adoption of Value Stream Management is critical in 2020. It will enable you to automate the outputs from Value Stream Mapping for ongoing progress monitoring. This lets a team connect all parts of the ever-complex DevOps toolchains with system-derived data based on cycle time. Teams that adopt Value Stream Management in 2020 will be able to base their next improvement experiments on data-driven decisions and prioritizations.
[ Read also: DevOps terms: 10 essential concepts, explained. ]
5. Tool fatigue will worsen before it gets better
The number of tools and frameworks in technology is daunting. The challenges IT teams face to understand, interconnect, and apply these will continue, and in 2020, there is no real resolution in sight.
The competition in the DevOps toolchain is fierce and flourishing.
The competition in the DevOps toolchain is fierce and flourishing. Events and conferences are filled with technology and best-practice sessions. Books, blogs, and videos are flooding email inboxes, and thought leaders are eager to share their expertise. Additionally, more open source tools are emerging to integrate new technologies.
To survive the challenges of complexity, it’s becoming increasingly important to have an automation strategy. As you work to develop this, don’t lose sight of the actual problems you are trying to solve and how you can get there by leveraging your own teams.
6. DevOps will become more measurable and metrics will become better defined
“What gets measured gets managed.” This quote is still valid, more than 60 years after Peter Drucker referenced it in his book, “The Practice of Management.” What we all want to avoid, however, is measuring for metrics’ sake.
Avoid the syndrome of measuring for metrics’ sake.
For the next few years of DevOps, the continuous improvement metrics we know today will continue to be the key metrics that matter. In 2020, we expect more organizations will agree on what to measure and adopt these metrics. Those looking for support can lean on the performance metrics outlined in research from DevOps Research and Assessment (DORA), which cites five measures of Software Delivery and Operational Performance (SDO) that can be used as leading indicators of success for high-performing DevOps teams. Key benchmarks within the report provide guidance on areas teams must improve in 2020. Selecting these key metrics and populating them with data will provide insights into the value and journey of DevOps.
7. DevOps teams will earn more and experience increased job satisfaction
Employees who continue down the DevOps path in 2020 will see benefits that go directly into their wallet and their job satisfaction. Automation will allow employees to work on more value-adding work instead of mundane manual tasks, resulting in improved job satisfaction and reduced stress levels.
In particular, DevOps engineers who see their work in automating and collaborating for improved software delivery can expect their salary to be significantly higher than peers in traditional roles, such as system administrators. Investments in training and certifications will also positively impact the quality of code and therefore may improve the business results. This, in turn, might finally change the value equation within more organizations and give IT more strategic seats at the table. It will certainly improve how other functional areas work with IT now and in the future.
8. There just might be service management and DevOps reunification
With the recent release of ITILv4, 2020 will be an interesting year for organizations that have been adopting DevOps and service management frameworks. The development and management of software products needs agile techniques, with a focus on value co-creation in a way that reduces waste. DevOps, service management, and other best practices like SRE can coexist to align teams, meet stakeholder demands, and improve the value delivered. Because digital transformation is not achieved instantly across an organization, established companies should begin with best practices and methodologies that are suited to their needs by starting small – then learn, build expertise, and scale up.
[ Read also: DevOps vs. ITIL 4 vs. SRE: Stop the arguments. ]
9. A new generation of IT takes over
The number of people who remember the days before DevOps is starting to shrink. The younger generation of workers on today’s IT and DevOps teams don’t remember the strict silos, with clear lines around areas of responsibility such as infrastructure, operations, application design, development, testing, and security. They don’t remember how this caused a lot of transition work between teams and groups. They don’t know that product owners, business analysts, architects, developers, testers, release managers, system administrators, and infrastructure owners had to agree and coordinate on the planning, development, testing, deployment, operation, and management of a piece of software. Just typing that sentence was exhausting; imagine living it.
The number of people who remember the days before DevOps is starting to shrink.
When we celebrated the 10-year anniversary of DevOps in 2019, we saw the wall removed between Dev and Ops. This is a reason for all of us to celebrate, and in 2020, expect to see this happen as more and more organizations adopt DevOps.
10. Artificial Intelligence adoption will rise, and DevOps teams should check it out
AI and machine learning (ML) were recently rated as the most important enterprise technologies of the next decade, according to a recent report from ISACA. Both will play a critical role in next-generation IT operations and DevOps teams. AIOps will give DevOps teams the ability to analyze more data faster, allowing them to improve key processes, tasks, and decision-making. In 2020, expect to see more DevOps teams adopting these tools, which automate the ingestion of fast volumes of data, use ML to analyze the data, and have the ability to leverage knowledge for automation or decision-making.
The AIOps market has already gained momentum, with 22 percent of IT enterprises using ML and data sciences as part of their work. The vendor landscape is broad, with a variety of leaders. This year, the AIOps market will continue its shift from a science project to the pilot and experimental stage.
MORE ON DEVOPS
Agile vs. DevOps: What’s the difference?
How to explain DevOps in plain English
3 DevOps, agile, and scrum myths, debunked
Much work remains in 2020 to aid in the future of DevOps adoption. It will require more than focusing on tools and techniques. DevOps initiatives must be registered as change programs, requiring a time, resource, and priority commitment from all business leaders. We’re hopeful that some of these trends will be realized this year, and that DevOps will be recognized as a new way of working.