This article summarizes findings originally reported on searchsoftwarequality.techtarget.com, based on a DevOps survey published by LogiGear Corp. in mid-2017. Supporting data on tooling popularity comes from StackShare and Packt’s annual Skill Up survey (also 2017).
The Core Problem: DevOps Still Means Different Things to Different People
Software testing toolmaker LogiGear Corp. released a DevOps survey in mid-2017, and one respondent’s comment cut to the heart of a persistent industry problem:
“Agile and Scrum were concrete processes that led to tangible and positive results in the workplace. On the other hand, I have yet to understand the term DevOps, except as a meaningless buzzword that has had zero effect on our work processes.”
This sentiment reflects a genuine challenge: unlike Scrum, which ships with defined roles, ceremonies, and artifacts, DevOps resists a single prescriptive definition. It encompasses culture, tooling, automation practices, and organizational structure — making it harder to adopt in a consistent, measurable way.
Key Survey Findings: Automation Is the Central Pressure Point
Among teams already operating in a DevOps environment, the survey surfaced several clear trends:
- 60% of DevOps practitioners reported feeling significant pressure to automate more of their pipeline.
- Nearly 50% flagged environment or test data problems as a major ongoing challenge.
- 60% said they were running more tests since adopting DevOps practices.
- 32% said they were running many more tests than before.
The increased volume of testing is a natural consequence of shifting left and integrating continuous integration (CI) pipelines — more frequent code commits trigger more automated test runs. But the data also makes it clear that the infrastructure and data management side of testing hasn’t kept pace: environment instability and test data availability remain friction points that automation alone does not solve.
Non-DevOps Teams: A Different Baseline
For teams not yet attempting DevOps, the picture was notably different:
- Only 46% felt strong pressure to automate — a meaningful gap compared to their DevOps counterparts.
- Just one-third reported that their existing Agile and Scrum practices were in good shape.
This suggests that the pressure to automate isn’t simply an industry-wide trend — it’s amplified by the continuous delivery demands that DevOps adoption brings. Teams that haven’t made the shift yet are under less pressure, but may also have less visibility into quality problems that faster release cycles would expose.
A Positive Signal: Ops and Test Teams Are Collaborating More
Not all findings pointed to friction. The survey included some encouraging data on cross-functional collaboration — historically one of the weakest areas in software delivery:
- 25% said their ops teams were very helpful when it came to running tests or supporting testers directly.
- 37% said ops teams regularly helped create and maintain stable test environments.
While these numbers show there is still significant room for improvement, they do indicate that the cultural shift DevOps requires — breaking down the wall between development, QA, and operations — is happening in a meaningful subset of organizations.
Popular App Hosting Tools in 2017: What StackShare’s Data Showed
Alongside the DevOps survey findings, StackShare’s real-time tooling index (as of June 2017) gave a snapshot of which application hosting tools were most widely adopted and most in-demand by employers.
Most Used (by Stack Count)
- NGINX — appeared in 11,200 stacks
- Apache HTTP Server
- Amazon EC2
- Heroku
- Microsoft IIS
- DigitalOcean
- Socket.io
- Firebase
- Apache Tomcat
- Google App Engine
Most In-Demand by Employers (by Job Posting Count)
- Amazon EC2 — approximately 2,790 job postings
- NGINX
- Microsoft Azure
- Apache Tomcat
- Heroku
The divergence between popularity in the developer community and employer demand is worth noting. EC2 topped the jobs list despite ranking third in stack count, reflecting the weight that enterprise cloud infrastructure carries in the hiring market. Azure’s jump into the employer top five — despite not appearing in the community top ten — points to the growing enterprise adoption of Microsoft’s cloud platform at that time.
Skills Developers Wanted to Learn: Docker Led the Way
Packt’s third annual Skill Up survey, which polled approximately 5,000 developers and tech professionals worldwide, shed light on career trajectories and learning priorities.
On compensation, C-level executives and CIOs ranked highest, while roles such as big data engineer, security engineer, and information architect earned above-average salaries. Geographically, US-based respondents reported the highest average salaries; South and Southeast Asia reported the lowest. Game, web, and mobile developers, along with technical support staff, ranked at the lower end of the pay scale.
The most revealing data point, however, was which technologies developers most wanted to learn:
- Docker — container technology; the clear top priority
- Python — general-purpose and scripting language
- Angular — front-end JavaScript framework
- Visual Studio — Microsoft’s integrated development environment
- Jenkins — open-source automation and CI/CD server
Docker’s position at the top of this list — ahead of programming languages and IDEs — underscores just how central containerization had become to the DevOps conversation by mid-2017. Jenkins’ presence on the list is directly tied to the automation pressure highlighted in the LogiGear survey: teams adopting DevOps needed CI/CD tooling, and Jenkins was the dominant open-source option at the time.
What This Means for DevOps Teams
Taken together, these survey results paint a coherent picture of the DevOps landscape in 2017:
- Automation demand is real and growing, but it is not evenly distributed — teams actively practicing DevOps feel the pressure most acutely.
- Tooling clarity matters: the gap between what developers use and what employers pay for highlights the importance of understanding enterprise priorities, not just community trends.
- Definitional confusion slows adoption: organizations that can’t clearly articulate what DevOps means for their context will struggle to implement it meaningfully.
- Test environments and data management are underappreciated bottlenecks — automation pipelines break down when the environments they run against are unstable or poorly maintained.
- Cross-functional collaboration between ops and QA is improving, but remains far from universal.
While this data is from 2017, many of these challenges — automation pressure, tooling fragmentation, and cultural resistance — remain recognizable to DevOps practitioners today.