Artificial intelligence is said to be driving intelligence in businesses and is doing the same for IT stores. For example, AIOps (artificial intelligence for IT operations) apply AI and machine learning to stream data from IT processes, screening for noise to detect, highlight, and fix problems.
AI and machine learning are also finding a common home in another emerging IT area: assisting DevOps teams to ensure the feasibility and quality of software that is moving at an increasingly rapid pace within the system and into the future. user hand.
As found in a Recent surveys get out GitHub, development and operations teams are turning to AI in a big way to smooth code flow through the software testing and evaluation phase, with 31% of the team actively using AI and ML algorithms for code review – more than double last year’s numbers. The survey also shows that 37% of the team use AI/ML in software testing (up from 25%) and 20% plan to introduce it this year.
Also: Understand Microsoft’s grand vision for building the next generation of apps
An additional survey of Techstrong Research and Tricentis confirm this trend. The survey of 2,600 DevOps practitioners and leaders found that 90% favor bringing more AI to the test phase of DevOps flows and see it as a way to address the skills shortage they are currently experiencing. must face to face. (Tricentis, a software testing vendor, clearly contributed to the results. But the data is important because it reflects a growing shift toward more autonomous DevOps approaches.)
There is even a paradox emerging from the Techstrong and Tricentis research: Businesses need specialized skills to reduce the need for specialized skills. At least 47% of respondents say that the main benefit of AI-powered DevOps is reducing skills gaps and “making it easier for employees to perform complex tasks.”
Also: DevOps nirvana remains a distant goal for many, survey shows
At the same time, the lack of skills needed to develop and run AI-powered software tests has been cited by managers as one of the top barriers for DevOps using AI, at 44 %. This is a vicious circle that will hopefully be overcome as more and more professionals enter training and education programs focused on AI and machine learning.
Once AI starts to be introduced into IT websites, it will help create a shift in process-intensive DevOps workflows. Nearly two-thirds of managers in the survey (65%) say that functional software testing is a great fit and would greatly benefit from AI-enhanced DevOps. “DevOps success requires large-scale test automation, generating large amounts of complex test data, and requiring frequent changes to test cases,” the survey’s authors said. shown. “This fits perfectly with AI’s ability to identify patterns in large data sets and provides insights that can be used to improve and accelerate testing.”
Also: Artificial intelligence projects grew 10 times over the past year, survey says
Along with possibly reducing skill requirements, the survey also identified the following benefits to bringing more AI into DevOps:
- Improve customer experience: 48%
- Cost reduction: 45%
- Increase the efficiency of the developer team: 43%
- Increase code quality: 35%
- Troubleshooting: 25%
- Release speed increase: 22%
- Coding knowledge: 22%
- Disability prevention: 19%
Early adopters of AI-enhanced DevOps tend to come from larger organizations. This is not surprising, given the larger concerns there will be more developed DevOps teams and greater access to cutting-edge solutions like AI.
Also: It’s time for tech teams to find their voices in the customer experience
Techstrong and Tricentis authors point out: “In terms of DevOps, these mature companies are marked by the progress they have made in streamlining their software development capabilities within 5 to 7 years. years as well as their finishing and finishing processes and pipelines”. “These DevOps organizations are cloud-based and use DevOps workflow pipelines, tools, automation, and cloud technologies.”
In the long run, using AI to support important aspects of DevOps is a smart idea. The DevOps process, for all its collaboration and automation, is only getting more tiresome as software is expected to fly out the door at an increasingly rapid pace. Leave it to the machines to handle a lot of the difficult aspects, such as testing and monitoring.
