During a recent webinar series on “Building a DevOps Culture & Infrastructure for Success,” we asked the audience to rate their confidence with knowing what features and/or fixes are in any given release. We were surprised to find out that only 12% were really sure that they could describe the functional changes within any given release with precision.
To help us further understand and overcome the barriers to realizing the vision of DevOps, we recently surveyed enterprise IT leaders to get their insights on DevOps adoption.
DevOps Adoption Moving Into the Mainstream
Approximately 73% of the respondents are currently using DevOps for production systems, pilot programs, or they plan to adopt DevOps in the next 24 months.
#1 Driver for DevOps – Improve Quality, Consistency & Repeatability
Historically, the need to increase deployment frequency has been cited as the primary factor driving DevOps initiatives. However, this seems to be changing. In our survey, the desire to improve quality, consistency and repeatability was the highest rated DevOps driver (88%). The need to increase deployment frequency has dropped to the second most common driver (62%) followed by the need to reduce failure rate of new releases (57%). As DevOps practices move further into the enterprise, increasing the overall quality of software delivery may be overshadowing the need for speed.
Only 33% of the respondents said that DevOps has been successful, while 54% said that DevOps has been moderately successful, and 13% said that it has not been successful.
Need to Improve Ability to Track the Flow of Business Value
Perhaps one of the biggest takeaways from our DevOps survey was the need to increase the overall organizational proficiency of tracking the flow of business value – from idea to production. Approximately 88% of the respondents gave their organization a moderate or low efficiency rating for their ability to track and manage features and/or fixes.
Part of the challenge may be the number of systems that need to be accessed in order efficiently manage and track features and/or fixes. Nearly 85% of the respondents are reconciling multiple systems to identify the business work items included within a specific environment or release at any given point in time.
Cumbersome Manual Processes Add Extra Effort
In addition, 87% said that pulling a list of features and/or fixes is manual – either very manual with spreadsheets, etc. (29%) or partially manual by combining and/or aggregating data generated by automated tools (58%).
Challenges Due to Disconnected and Fragmented Delivery Tools
So what are some of the challenges that organizations have experienced due to the disconnected and fragmented delivery tools? Here are some of the specific quotes from the survey respondents:
“Missed priorities, missed opportunities and rework”
“Delayed delivery and compromised quality”
“Bad code, bad data, no change management, lack of understanding”
“No traceability and no one knows what is in given releases, raising necessary questions”
“Manual deployment of the features has a high margin of error”
“Unknown and untested code making it to production. Incomplete functionality being delivered.”
“Re-work, extra effort, release delivery risks”
“System in off line for many minutes”
“Functional outages and degraded performance due to not deploying an integrated set of changes – code, database, and configurations – with specific features”
“Conflicts between delivery teams working on related (known or unknown) systems”
“Lack of confidence about possible errors”
Introducing Unified Software Development and Delivery
In my last blog post, I explained that while propagating the vision of DevOps has been a success, executing against it often remains challenging – especially in the enterprise. I believe that successfully unifying plan, develop, validate, deploy and run workflows is still challenging for two fundamental reasons:
- Plan and develop work items (features, fixes, stories, etc.) are not directly linked to operational outputs (builds, artifacts, environments, etc.)
- Lots of fragmented automation make it difficult to orchestrate and creates many pockets of siloed data.
In order to make the vision of DevOps a reality, a truly unified platform that supports the end-to-end delivery stream – from idea to production – is a primary requirement. The VersionOne® Continuum™ for DevOps solution is one example of this type of platform. For more information, visit https://www.versionone.com/product/devops/