The Current State of Continuous Deployment

Less than half of software development professionals are using continuous integration (CI) or continuous deployment (CD), despite the fact these processes improve reliability and deployment velocity. According to new data, teams recognize the value of fast, reliable releases but encounter challenges with excessive manual tasks and process complexity.

Armory and Gartner Peer Insights explored the current state of continuous deployment by surveying 400 engineering and operations professionals and compiling three reports. Here’s a look at the findings.

Continuous deployment in app development

Researchers analyzed respondents’ answers to understand the current state of deployment reliability, levels of CI/CD adoption and barriers to innovation.

Automation drives continuous deployment adoption. The survey revealed 48% of respondents are motivated to use continuous deployment to lower change failure rates and decrease mean time to restore by automating debugging and monitoring tools, while 41% value the cost savings realized by reducing manual tasks.

Why are more teams not making the investment? The majority (71%) cite existing legacy software as the most significant cost barrier to deployment innovation.

More than 80% of surveyed professionals rate reliable deployments as their top priority. Google’s 2022 State of DevOps Report found reliability to be the fifth key to improving organizational performance (the other four being DORA metrics). Your deployment velocity does not matter if customers cannot use the update. Despite this target, reliability and consistency are challenging for a third of respondents.

Overly complex deployments to multi-cloud environments rose to the top of app development and deployment issues that engineers must address, with 59% listing it as the most significant problem.

For those who adopted continuous deployment, 68% experienced the need to create automated tests for each new feature, bug fix or improvement. 

Enterprise scale deployments

The second report sheds light on organizations’ efforts to achieve speed-to-market.

The most common deployment cloud environment is a multiple public and one private cloud setup, used by 45% of respondents. At least half of IT leaders cite Amazon Elastic Kubernetes Service (EKS), virtualized environments/virtual machines (VMs) and Azure Kubernetes Service (AKS) as current deployment targets. Teams most commonly use rolling deployments (60%); only a third use canary methods.

How do development professionals measure software quality? The top three metrics are:

  • Customer satisfaction (67%).
  • Competitive response (62%).
  • Financial results and return to stakeholders (60%).

Scaling remains a challenge, with only a quarter of respondents feeling their organization can reliably manage hundreds of apps or microservices.

Software deployment pain points and risks

The final report found development teams’ top pain points are too many manual steps, lack of consistency and overly complex deployments to multi-cloud environments. The majority blame poor processes and workflows, developer talent or their tech stacks for the difficulties.

One time-consuming task frequently done manually: monitoring application health metrics. Nearly half of developers use this method, and only 37% use automated canary analysis.

Recognizing the importance of avoiding downtime, 64% of respondents rate tightly coupled services that can deploy and roll back together as crucial to their organization.

Room for improvement

These survey results paint a picture of the state of the industry. Release velocity and reliability are more crucial than ever while development complexity builds. Continuous deployment can solve or improve many common pain points and help achieve elite status on core KPIs, yet adoption remains slow.

Reliable deployment ties closely to revenue generation and customer satisfaction, helping DevOps leaders build a compelling economic case for continuous deployment adoption to improve metrics, minimize downtime and generate higher-quality software. Teams who have already implemented automation are immediately seeing the benefit of a streamlined and less burdensome process. As most businesses look to scale, they need dependable solutions to generate reliable deployments that drive business value and provide a competitive advantage.




Source link