How AI Will Change DevOps

Share twitterlinkedinmail

Intelligent machines aren’t coming, they’re already here. The question is: how will they change things? We’ve already discussed the future of AI and ITSM,  but today I want to take a deeper look at how AI will impact DevOps.

Much of DevOps is about the automation of tasks. It focuses on automating and monitoring every step of the software delivery process. DevOps encourages enterprises to set up repeatable processes that promote efficiency and reduce variability. Artificial Intelligence (AI) and Machine Learning (ML) can help improve those efficiencies and automate even more of the process so DevOps practitioners can focus on bigger, more complex initiatives.

DevOps experts have a lot to gain by adopting AI and ML. According to ServiceNow’s report “The Global Point of View”,  85% of C-level executives believe AI can offer value in terms of accuracy and rapidity of decision making. 60% of C-level executives surveyed said decision automation can contribute to their organization’s top-line growth. But according to that same report, only 27% of have hired team members with skills in machine learning.

For current DevOps practitioners and IT leaders, it will pay off to start understanding how AI will change DevOps and how you can exploit these newer technologies

Data Accessibility

Perhaps the biggest impact AI and ML will have on DevOps is the capability to access and correlate data from disparate sources. DevOps activities generate large amounts of data. That data can contribute to many aspects of IT: streamlining workflows, monitoring systems, and diagnosing issues. However, the quantity of data can often become overwhelming for teams. So rather than looking directly at the data developers define tolerance thresholds and use only breaches of those thresholds as conditions for action. But by doing this, they are identifying only outliers and ignoring the majority. That can create larger problems because IT organizations are unable to see an informed, deep view with only outlier data.

AI can be used to collect data from multiple sources and prepare that data for evaluation. ML can then be used to identify and predict any alarming patterns and create recommendations based on those patterns. This helps to keep DevOps in a “predictive state” as opposed to a reactive one and provides continuous feedback throughout the process.

And even when there are alerts that may cause DevOps teams to be “reactive,” AI can help with that as well. Many DevOps teams may be accustomed to receiving a high number of alerts, but ML can help manage those alerts and prioritize them based on factors such as past behavior or the magnitude of the current alert. This way DevOps teams can continue to move quickly and efficiently and “fail fast,” as they are often encouraged to do.

Mask Operational Complexity

AIOps is an emerging application of AI and ML that helps DevOps teams have a consolidated and unified view of all components of the toolchain (and more). Using AIOps, an engineer can view all the alerts and relevant data produced by the tools in a single place and the team will have a holistic view of an application’s health. In many cases, the AIOps tool can take automated actions in response to data patterns and conditions, using ML and associated algorithms.

Improved Testing

While integration testing is done as part of trunk code updates, AI and ML can be used to perform deep integration and regression testing to identify potential poor development practices.

Even more automation

DevOps wants to automate as much as possible, but for many organizations, automation is focused on code deployment. Through the use of AI and ML, infrastructure configurations or builds can be automated, tasks within processes automated, processes orchestrated, as well as remediation responses to alerts.

Where to Start Adopting AI with DevOps

Before you rush out and invest in an AI or ML tool, you need to establish your DevOps foundation so that you’re ready and capable of handling the changes AI can bring to your organization.

Remember, AI can only do what it has been designed to do. Just as with ITSM, good AI will not fix poor DevOps practices. You can start preparing for adopting AI into your DevOps environment by reviewing three key factors first.

  • Processes – If your processes are not documented, clear or follow then AI can’t automate them! Start with a clear, documented process and then AI can step in to help it operate more efficiently.
  • Standardization – The more standardized the environment, the easier it will be to introduce AI / ML into the environment. Standardization reduces variably and integration challenges. Standardize not only on the infrastructure, but also standardize tools and APIs as well.
  • Map the complete value stream – Some DevOps teams only look at the IT value stream as including only software development and deployment, which are important, but not inclusive of everything that is done to deliver value to a business. The complete IT value stream must include not only operations and support, but QA, security, portfolio management, and the consumer.

The future of DevOps is bright. AI and ML can revolutionize how DevOps operates but your team needs to be primed and prepared to handle these changes prior to purchasing an AI tool. Additionally, working with your team to embrace AI will contribute to their career advancements as AI and ML continue to play larger roles in DevOps and IT as a whole.

Interested in learning about AI? Contact Tedder Consulting to learn about AI workshops and consulting.

Looking for DevOps training? Learn about our DevOps training here.

Share twitterlinkedinmail

Leave a Reply

Your email address will not be published. Required fields are marked *

logo