VictorOps is now Splunk On-Call! Learn More.
With every passing year, our understanding of DevOps deepens – leading to new techniques, tools and processes for bringing developers and IT operations closer together. And, as we discussed in 2019’s DevOps Trends, security continues to become more tightly integrated with development lifecycles and operations workflows. But, as we head into 2020, our sentiments are always changing, so it’s a great time to reflect on the way people are looking at DevOps, popular trends, and common questions coming up across the industry.
In 2019 we saw automation, artificial intelligence and AIOps adopted more often and researched by the DevOps community. Containers and functions-as-a-service continued to grow, allowing developers and operations teams to take greater control of their applications and infrastructure without hindering speed. Observability increased as APM tools, logging software, and traces/error tracking functionality became more robust. And, via continuous improvement, businesses, product teams and engineering teams generally bolstered collaboration across departments – leading to smarter workflow management and greater transparency into operations.
While most DevOps trends from 2020 still persist, let’s dive into the top five that appear to be newly developing in 2020.
By no means is this an all-encompassing list. But, through a little bit of research (mainly through Google Trends), we can see how people are looking at the topic of DevOps and the subject matter that’s most interesting to them. So, let’s take a peek under the DevOps hood to find out what’s changing in DevOps this year:
SysAdmins, IT security professionals and engineering managers are all continuing to better understand DevOps. There are fewer conversations about “DevOps Engineers” and more conversation about a holistic DevOps approach to engineering and the business. More developers and IT operations professionals are finding the nuance between a siloed DevOps team and a more collaborative approach to the development, deployment and operations lifecycles.
Instead of focusing on teaching developers all of the IT skills and teaching IT teams to learn programming languages, organizations are leaning into processes that help pull the greatest value out of everyone on the team. Engineers are learning to specialize in their field while building a broad understanding of the rest of their applications and infrastructure – leading to faster development processes and more resilient services.
As more teams adopt DevOps, more people begin to understand basic underlying technologies and skills. Practices such as unit testing and chaos engineering are becoming more common within engineering organizations, driving less confusion around the topics, and ultimately leading to fewer people searching for these terms on Google. More specific practices that were new five years ago are becoming old hat for IT practitioners and DevOps-minded software developers.
Engineers are more comfortable taking accountability for the code they write, often taking on-call responsibilities and responding to alerts in real-time for problems with the software they’re building and maintaining. People are becoming comfortable working with a little more ambiguity between the traditional “developer” and “operations” roles – leading to better collaboration and more buy-in for a DevOps culture.
While AWS is the clear market leader in cloud computing, storage and operations, Azure and GCP are also continuing to see amazing growth. DevOps-minded teams are moving to the cloud more and more in order to facilitate faster changes to production, improve visibility across all applications and infrastructure, and build more resilient architecture. Cloud architecture such as AWS, Azure or GCP can turn platforms, functions and infrastructure into simple-to-operate services (PaaS, FaaS and IaaS). By addressing scalable hardware problems with code, it’s becoming easier and easier for DevOps organizations to make changes quickly at the push of a button.
In many organizations, engineers will conflate the use of DevOps and site reliability engineering (SRE). However, SRE and DevOps are very different disciplines, and people are starting to recognize it. The knowledge and expertise of both functions are continuing to grow in 2020 and businesses are starting to recognize the value of them both. While DevOps is a mindset, a methodology for tightened collaboration between developers and operations teams, SRE is a more specific technical expertise held by engineers.
Site reliability engineers are tasked with writing code and applying software development solutions to IT operations and infrastructure problems. DevOps is a cultural shift toward greater transparency and collaboration from product development to testing to production upkeep. More and more businesses are learning how SRE and DevOps work together to drive resilient continuous deployment pipelines.
Contrary to what you may think, Google Trends shows a drop in search interest around continuous integration, continuous deployment and continuous delivery. This trend indicates that engineering teams are beginning to look at CI/CD as a requirement, not a nice-to-have. This drop in search interest also shows that more people are understanding the nuance between CI, CD and the other CD. Businesses are seeing that CI/CD can be used as a competitive advantage in order to serve value to customers faster and can also help DevOps-minded teams resolve incidents or rollback changes more easily.
A well-built CI/CD pipeline will not only improve velocity but it should help improve the overall reliability of the services supported. It’s great to see that 2020 should start to see the normalization of CI/CD and DevOps.
If you’re not already on the bandwagon for any of these top five trends for DevOps, maybe it’s time to make that your new year’s resolution. But, continuous improvement is at the core of DevOps and not all organizations are at the same level of engineering and IT maturity. So, if you’re not looking into any of these five trends, I still encourage you to try new processes, test some new tools and don’t be afraid of failure. Failure in the development process and in production environments is really the only true way to find out what works and what doesn’t for you and your team. Happy DevOps’ing this year – hope everyone’s new year is off to a great start!
Be one step ahead of your next incident and simplify your approach to incident response with VictorOps. Sign up for a 14-day free trial or request a free personalized demo to start making the most of automation and collaboration in the incident lifecycle.