VictorOps is now Splunk On-Call! Learn More.
Kelsey Loughman April 03, 2019DevOps
The DevOps train continues to gain momentum. More IT and software development teams are learning to shift left – running more security and reliability tests earlier in the development lifecycle. DevOps isn’t just a buzzword, it’s a way for teams to improve collaboration and workflow transparency – leading to the faster development of more reliable applications. Last year, we wrote about the top DevOps trends of 2018 – but now we’re here to look at any changes to DevOps trends in 2019.
If you look at Google Trends for the topic of DevOps, you can see the term has undergone a 900% increase in interest since 2013. But, if you look closer, you can even see DevOps interest has already increased by 30% from December 2018 to March 2019.
Image Source: Google Trends, “DevOps”
Not only are people showing more interest in DevOps at a high level but more people are adopting DevOps-related technologies and practices. According to a Hackernoon article citing Statista, there was a 7% increase in DevOps adoption from 2017 to 2018. And, even further, the DevOps software market is expected to grow from $2.9 billion in 2017 to $6.6 billion in 2022 according to estimates from IDC.
While automation has served as one of the core tenets of DevOps from the get-go, machine learning and AI-focused technologies are also starting to make their way into DevOps practices. AI and machine learning use cases are slowly being introduced across all aspects of the software delivery lifecycle. Computers can be used to identify bottlenecks, identify shortfalls in automation and predict failures in your application or infrastructure – sometimes before they even happen. Over time, the system can learn about itself and become smarter.
The key to leveraging AI and automation in 2019 is thinking about how they affect human workflows. In fact, Gartner conducted a study showing that at least 40% of new development projects will have AI co-developers on their team by 2022. Hybrid cloud, more complex architecture, microservices and ever-faster CI/CD processes require effective automation, machine learning and the rise in AIOps to make operations easier for the people on your team. DevOps teams in 2019 are really starting to push the envelope on automation and AI in incident management and software delivery.
Just look at the Google Trends chart showing the growth in interest for the search term, AIOps since 2014 (…and it’s predicted to grow):
Image Source: Google Trends, “AIOps”
If you ever attend a DevOps event, I’m willing to bet you’ll see at least one session about Docker Swarm or Kubernetes. In the past, the concept of microservices and containerized applications has been somewhat novel. But, the normalization of containers is starting to occur this year. In a forecast from 451 research, the application container tools market is expected to reach $2.7 billion by 2020. When dealing with complex systems at scale, it becomes easier to segment your teams into smaller squads and build your services around container orchestration technology.
As container technology becomes more mainstream, so does serverless computing, also known as FaaS. Functions-as-a-Service are helping teams create event-driven systems that only allow applications and tasks to run when you need them – reducing costs and creating operational efficiency. According to a survey from the Cloud Foundry Foundation, 22% of users are already using serverless technology – and nearly half of users are currently evaluating it. Both operations and software development is trending toward cloud and serverless functions in 2019, leading to deeper DevOps implementation.
DevOps principles don’t only apply to engineering teams. The more you spread DevOps between business units and individual teams, the better the entire organization will be at moving from product development to sale. Every year, teams are realizing that costs of downtime can amount to hundreds of thousands of dollars per year – from lost sales to negative outputs in operational efficiency. The faster engineering teams develop resilient applications and infrastructure, the faster you can enable business teams to sell the services.
With greater visibility into DevOps workflows and deeper collaboration across all business units, the whole team can quickly see where features are in the product roadmap and use the information to take action. In 2019, DevOps will be used more and more by teams to drive business value and reduce the number of works in progress (WIP) sitting stagnant in the delivery lifecycle.
Tightening relationships between developers and IT, creating more reliable applications and services and shortening the feedback loop between teams was the primary focus of DevOps. But, as teams have become better at implementing DevOps, people have started to shift security further left in the SDLC – adding security testing earlier into the software development and delivery lifecycle.
You can see the increased integration of security into DevOps workflows by looking at Google Trends for the term DevSecOps.
Image Source: Google Trends, “DevSecOps”
As you can see, people are thinking more about DevSecOps every year since 2014. Shifting security and QA testing left in your CI/CD pipeline helps everyone in the organization release reliable, secure software faster. In 2019, we only expect the integration of security, reliability and speed to become even tighter for DevOps teams.
Addressing unknown unknowns in your system and adding observability across the entire development and delivery lifecycle continues to gain importance in 2019. Deeper knowledge of the way your system behaves can help teammates not only build new services faster but also helps on-call teammates fix critical incidents faster. With more distributed systems and cloud-based applications and infrastructure, creating observability throughout the entire system becomes difficult. But, 2019 is sure to offer up more innovative tools and processes for DevOps teams looking to implement actionable observability in their technology stack.
As always, DevOps misconceptions and values will persist in 2019 and continue to evolve. But, the core philosophy of continuous improvement in DevOps will never go away. DevOps teams will continue to drive innovation and reliability from the beginning of software development to the upkeep of systems in production.
Whether you’re improving on-call processes, adding automation across the delivery lifecycle, or creating observability across your applications and infrastructure – the current DevOps trends are shaping up to make 2019 a great year for software developers and IT professionals everywhere.
Learn how teams are implementing DevOps to create a holistic system for software delivery and on-call incident management in our latest eBook, Why DevOps Matters. Improve your team’s quality of life by simultaneously releasing software faster and making on-call suck less.