How DevOps Is Progressing
How DevOps Is Progressing
Developers are getting more done automatically, so they're able to focus on higher value work.
Join the DZone community and get the full member experience.Join For Free
SaaS and greenfield organizations adopt the DevOps methodology from the beginning as one team. Many legacy enterprises are still dealing with silos but they are moving to the cloud and implementing DevOps.
DevOps in the cloud is about feedback. Organizations want to bring that knowledge to their development teams as quickly as possible.
There are many teams supporting legacy systems on mainframes as well. Companies are using DevOps and realize the need to get other teams to adopt across the organization.
A big change is AI/ML data coming from existing systems. Organizations are able to find patterns that enable them to cross-sell. They are not able to do that with traditional development.
Based on CA's annual survey adoption of DevOps is increasing as is DevSecOps with security becoming integral to DevOps. Veracode's research confirms that security is shifting left.
DevOps focuses on islands of automation. Now it’s going mainstream. How to piecemeal automation across the SDLC through CI/CD pipeline. DevOps improves code quality and speed to market. More and more customers are understanding this and want the benefits for their own organizations.
The third area of growth is around continuous testing. Not just doing testing as a box, embedding it from planning to production. Sprint telecommunications has been doing this for awhile.
Apple is seeing an increase in quality issues. They need to do end-to-end testing because of the number of platforms there are on. When you focus on introducing new features you need to continually test and run CI/CD.
Test the right things and see what's changed. Every software has a million use cases. Organizations need to provide information to data scientists to solve specific problems like mean-time to resolution (MTTR).
CA can help show organizations how to use their data and improve their performance. They provide a data scientist in a box.
DevOps are becoming more AI based -- how do I get the right data and learn what’s happening with customer experience (CX) to improve on it? Applications will learn with more data. Organizations must be disciplined to collaborate even more important in an AI world. ML is basically computational statistics but it must have the right data.
Take recommendation engine efficacy from 60 to 70% to greater than 90% to see the ROI. If you're not using data better than humans, you're wasting your money.
Developers initially thought they'd be wearing a pager in a DevOps environment but they quickly realize the benefits of getting feedback sooner and seeing how the application performs before deploying. Developers are getting more things done automatically so they can focus more on creative things.
Phenomenal changes in cultural transformation are necessary. DevOps is critical to increase velocity, reduce errors, improve quality, and reduce costs. It's now being embraced by key stakeholders.
The CTO articulates the vision and enables teams to reach them. Spotify uses squads, chapters, tribes, and guilds to accomplish this. ING has adopted the Spotify model. The idea is to cut through the hierarchy and enhance collaboration. More organizations see value by bringing development teams on board more quickly.
Opinions expressed by DZone contributors are their own.