To succeed in DevOps these days, go hybrid

Survey identifies top skills needed for the next phase of DevOps: process skills and knowledge, ability to automate, CI/CD
Written by Joe McKendrick, Contributor on

DevOps is not just a technical undertaking, it's a business proposition. This calls for hybrid skills that enable a more holistic view of the entire software development and deployment process. With a majority of enterprises (52%) intending to ramp up their recruiting of DevOps skills. close to two-thirds, 65%, reported having difficulties with finding these combinations of skills.

Photo: Joe McKendrick

That's the takeaway from a survey of 1,260 IT managers by the DevOps Institute, which finds process skills and knowledge are now the top skills sought for DevOps initiatives, surpassing automation skills, the leading skill sought in last year's survey. In addition, DevOps adoption, Agile, and site reliability engineering (SRE) all have gained percentages as must-have skills. 

In the first stage of DevOps adoption, "users were learning about the technology and its potential benefits and implementation difficulties," the survey's authors point out. "The next logical stage is to establish a process and rules through which the technology will be used."  

The DevOps journey is still very difficult for more than 50% of respondents. "DevOps is a fundamental change in the traditional structure of IT. It not only represents the adoption of new technology, but also an organizational transformation challenge with all that it implies with the 'evangelization of the existing responsibility silos. I certainly may be perceived as threatening for several structures which are comfortable in the existing organizational model."  

DevOps topologies primarily used to today are a huge challenge. The survey's authors identified three distinct models: The one that is leveraged the most is the DevOps team silo or the DevOps advocacy model where there is a separate DevOps team between dev and ops. The second most-used mode is the DevOps collaboration model where Develops is everyone's job, and the third most-deployed team model is that of a DevOps tool team where the DevOps team is responsible for the tooling required.  

Enterprises increasingly seek hybrid skill sets to move forward with DevOps, the survey shows. "The ideal candidates for a DevOps role at this stage demand a holistic view and understanding of issues face by developers, ops teams, and the possibilities of automation." That's because "human, process, knowledge and automation skills are equally important."  The hybrid DevOps specialists leverages skills from a wide area of fields which range from technology skills such as cloud infrastructure, to functional skills such as IT operations, to processes and framework skills such as SRE or Agile."

The following skills top the DevOps Institute survey's list:

Top DevOps Hybrid Skills

  • Process skills and knowledge  68%
  • Automation skills 67%
  • Soft skills   61%
  • Functional knowledge (networks, storage, etc.)  36%
  • Specific automation tools    29%
  • Business skills   23%

Top DevOps Functional Skills

  • IT operational knowledge   52%
  • Security practices   52%
  • IT infrastructure knowledge   50%
  • Application development and design   45%
  • Quality assurance knowledge   32%
  • Business continuity/disaster recovery knowledge  31%
  • Testing knowledge  31%
  • Network knowledge   28%

Top DevOps Technical Skills:

  • CI/CD tool chain skills    69%
  • Cloud platform and cloud environment knowledge  63%
  • APIs    41%
  • Analytical knowledge   31%
  • Multiple programming languages   28%
  • Experience with UI, web and middle tier services  28%


OctoML CEO: MLOps needs to step aside for DevOps

OctoML CEO: MLOps needs to step aside for DevOps

AI & Robotics
Three ways companies are changing how they think about DevOps
Mid adult Chinese man sitting at desk using computer, IT supporter assisting office worker, software developer looking at screen and concentrating

Three ways companies are changing how they think about DevOps

Fair forecast? How 180 meteorologists are delivering 'good enough' weather data

Fair forecast? How 180 meteorologists are delivering 'good enough' weather data

Big Data