Top 2020 Trends in Cloud computing, DevOps, and Infrastructure.
Fueled by Enterprise IT professionals moving beyond legacy practices to embrace new process and tools with their Cloud / Hybrid IT transformation, and younger organizations native adoption of containerized micro services and service orchestration technologies, the DevOps tools market is hitting an incredible growth, with IDC stating progression figures from $5.2 billion in 2018 to a whooping $15 billion by 2023, and the public IaaS and PaaS markets doubling in size in just two years.
Spurred by this very dynamic context, the development pace of new DevOps tools might seem impossible to follow. So here’s a big picture of the key trends emerging in the Cloud computing industry for the year 2020 and beyond, compiled from a series of analyst articles.
Micro-services architectures and containers technologies continue their way towards becoming mainstream.
The need for more flexibility and shorter release cycles will not slacken in the years to come which is why O'Reilly predicts that micro-services architectures will be featured in 90% of all applications by 2022.
Encompassing the dreaded need to refactor some legacy systems too, the quest to achieve a better operational efficiency will prevail with 63% of enterprises adopting microservices architectures next year, according to IT Pro Portal.
Kubernetes wins the day and favors ancillary services hatching around it.
As a logical outcome of the emergence of containers technologies, the use of orchestration and containers services such as Kubernetes, EKS, AKS and GKE is on the rise too.
Beyond this obvious statement, what is interesting to note is the steep progression of what is being called « Balkanization » of Kubernetes companion technologies by CRN. Indeed while new silos of satellite and non-CNCF (Cloud Native Computing Foundation) incubated projects are created and largely promoted around Kubernetes, odds are that the CNCF will maintain a tight grip on it's own roadmap and keep on incubating new tools to round out the stack in the years to come.
There is no questioning the fact that Kubernetes has become the standard for container orchestration. While the impressive figures reported by O'Reilly of a 2,300% increase in KubeCon attendance (from 500, skyrocketing to 12,000 people in 2019) might still be attributed to some kind of crazy hype, impacts on the job market are crystal clear : a 2019 national (USA) job search on LinkedIn turned up 16,744 open positions for Kubernetes-related roles confirming thus the progression of Kubernetes adoption and all directly related projects as a solid trend of 2020 and beyond.
Multi-Cloud gains ground and calls for specific skills and tools.
Fueled by the need to avoid vendor lock-in and the ability to leverage the best services from each cloud provider, plus the fact that applications become even more portable and compute cycles easier to procure in real time, multi-cloud truly bloomed in 2019 and should grow in 2020, so DevOps.com. Most studies suggest that more than 80% of enterprises now have a multi-cloud strategy, Amazon Web Services (AWS), Microsoft Azure and Google Cloud being obviously among the most commonly used cloud providers.
Surprisingly enough, this trends colludes with another one, also stated by DevOps.com : the rise of the "Enterprise Cloud Manager". Largely supported by online certifications programs delivered by most cloud providers, more and more team members now state the name of their cloud provider – such as AWS – in their job title, which clearly shows an increased individual expertise to manage and architect a cloud specific infrastructure.
But, as organizations of all size increasingly adopt several Clouds for their workloads, the need for consolidated multi-cloud tools rises. Those tools are not supposed to only address management and control purposes but are also meant for operational actions to be taken across multiple clouds and to simplify operations as native cloud providers consoles are not always easy to use and will not provide aggregated functions.
AIOps gains ground
2020 should see many organizations starting experimenting with some level of machine learning.
While mainstream adoption of AIOps still isn’t breaking out, introducing some degree of artificial intelligence to solve enterprise specific issues (like intelligent alerting, root cause analysis, threat detection, or predictive analytics), and increase overall operational efficiency will prove crucial and thus, IT professionals, DevOps and SRE teams should note a clear progression in the ecosystem of AIOps tools vendors.
Shift left on security
Though maybe still a buzzword, DevSecOps seems to be gaining ground with more and more organizations resorting to this set of best practices and looking for security tools to integrate within their software development workflow. But although the days of postproduction testing seem to be fading, there’s still some ground to cover, so O'Reilly.
Indeed, a 2019 Symantec survey on cloud security threats reports that 65% of businesses in the cloud still aren’t using multifactor authentication. The same survey also notes that an astounding 85% of Symantec’s cloud customers aren’t employing security best practices. Both figures clearly illustrating the progression margin within the sector, there is an obvious need to crank up the efforts on compliance, security tools and process implementation in the years to come.
More automation calls for smarter automation and traceability
As container usage grew along with orchestration technologies, so did complexity. Struggling to face all the tasks at hand, organizations have taken an opportunistic approach to automation, resulting in automation maturity that vary significantly between teams, some of them already resorting to various DevOps automation tools like Terraform or Ansible, and others pretty much left behind.
Gartner's Top 10 Trends Impacting Infrastructure & Operations for 2020 predict that By 2025, over 90% teams will have an automation architect, up from less than 20% today, providing for a clear automation strategy to guide choices relevant with the organizations’s use cases and strategy.
Along with this move towards automation and smarter automation choices, traceability should gain significant attention in 2020, so IT Pro Portal. Indeed, Continuous integration, continuous deployment (CI/CD) and release automation via DevOps tools also provides solid extensibility through webhooks that can be leveraged for traceability, making it a fast-moving trend for 2020.
As DevOps teams progressively move to Continuous Infrastructure Automation, making constant changes to multiple parts of a workflow at the same time, organizations will increasingly need to to trace the full lifecycle of their DevOps workflow in an effective way.
In 2019, container usage grew along with other models such as serverless and PaaS that allow developers to worry less and less about infrastructure, scaling and availability and focus on what they like most : run code and add functionalities to their applications.
Catching up with Amazon AWS that opened the game in 2014 with the first release of AWS Lambda, Google Cloud’s Functions, Microsoft’s Azure Functions, and other cloud providers are now offering their serverless solutions.
More and more companies now adopt a serverless-first approach after a basic lift and shift move of some of their legacy systems to the cloud. This and the fact that some startups also leverage the cost effective / agile application development benefits of resorting to serverless, confirm the fact that, like microservices, serverless will see an upsurge in usage as the age of the monolith fades.
Scaling DevOps agility
As DevOps best practices spread throughout organizations, finding the right approach to successfully scale all operations is no easy process.
In it’s "Top 10 Trends Impacting Infrastructure & Operations for 2020" report, Gartner clearly states "We believe that the vast majority of organizations that do not adopt a shared self-service platform approach will find that their DevOps initiatives simply do not scale. Adopting a shared platform approach enables product teams to draw from an I&O digital toolbox of possibilities, all the while benefiting from high standards of governance and efficiency needed for scale. By 2023, 90% of enterprises will fail to scale DevOps initiatives if they do not create a shared self-service platform. A shared self-service platform provides a “digital toolbox” of I&O capabilities to help multiple DevOps teams create, release and manage products, while ensuring consistency and streamlined efforts at scale ».
Beyond those trends showing a unique growth on the market, there is also some degree of maturity in the adoption of containerized micro services and service orchestration technologies.
Gartner's declaration on the late 2019 #GartnerIO conferences series stating that "Automation is important to infrastructure as blood is to the body. Orgs with it are predicted to have 3x customer satisfaction numbers as those without it" clearly underlines the fact is that the adoption of DevOps tools and automation has now become a key differentiating item on a business level.
Cloudskiff is the first Infrastructure-as-Code industrialization platform for Kubernetes built for the multi-cloud age.
Simplifying the process of delivering secure cloud infrastructures, we provide DevOps with a collaborative tool that improves the end to end process of deploying cloud native applications on multiple clouds. CloudSkiff manages the lifecycle of your infrastructure stack with clean Infrastructure-as-Code, so that you can focus on building and running your apps!
OReilly : Infrastructure and Ops Newsletter
Devops.com : The 6 Cloud computing trends of 2019
IT Pro Portal : 6 DevOps predictions for 2020