Last year, Australia’s two largest supermarket chains suffered nationwide technical issues, forcing the companies to close stores until the issues were fixed. The result: Lost revenue and frustrated customers. The reality is that IT teams are dealing with increasing amounts of data and a variety of tools to monitor that data, which can mean significant delays in identifying and solving issues.
“ The long-term impact of AIOps on IT operations will be transformative”
“IT operations is challenged by the rapid growth in data volumes generated by IT infrastructure and applications that must be captured, analyzed and acted on,” says Padraig Byrne, Senior Director Analyst at Gartner. “Coupled with the reality that IT operations teams often work in disconnected silos, this makes it challenging to ensure that the most urgent incident at any given time is being addressed.” To prevent, identify and resolve high-severity outages and other IT operations problems more quickly, businesses are turning to artificial intelligence (AI) for IT operations (AIOps).
What is AIOps?
Put simply, AIOps is the application of machine learning (ML) and data science to IT operations problems. AIOps platforms combine big data and ML functionality to enhance and partially replace all primary IT operations functions, including availability and performance monitoring, event correlation and analysis, and IT service management and automation. AIOps platforms consume and analyze the ever-increasing volume, variety and velocity of data generated by IT and present it in a useful way.
“ IT leaders are enthusiastic about the promise of applying AI to IT operations”
Gartner predicts that large enterprise exclusive use of AIOps and digital experience monitoring tools to monitor applications and infrastructure will rise from 5% in 2018 to 30% in 2023. According to Byrne, the long-term impact of AIOps on IT operations will be transformative. “IT leaders are enthusiastic about the promise of applying AI to IT operations, but as with moving a large object, it will be necessary to overcome inertia to build velocity,” says Byrne. “The good news is that AI capabilities are advancing, and more real solutions are becoming available every day.”
How to launch an AIOps initiative
- Don’t wait. Become familiar with AI and ML vocabulary and capabilities today, even if an AIOps project isn’t imminent. Priorities and capabilities change, so you may need it sooner than you expect.
- Choose initial test cases wisely. Transformation initiatives benefit from starting small, capturing knowledge and iterating from there. Take the same approach to incorporating AIOps for success.
- Develop and demonstrate your proficiency. Demystify AIOps for your colleagues and leadership by demonstrating simple techniques. Identify skills and experience gaps, then assemble a plan to fill those gaps.
- Experiment freely. Although AIOps platforms are often products of substantial cost and complexity, a great deal of open-source and low-cost ML software is available to enable you to evaluate AIOps and data science applications and uses.
- Look beyond IT. Leverage data and analytics resources that may already be present in your organization. Data management is a huge component of AIOps, and teams are often already skilled. Business analytics and statistical analysis are key components of any modern organization, and many techniques span problem domains.
- Standardize where possible, modernize where practical. Prepare your infrastructure to support an eventual AIOps implementation by adopting a consistent automation architecture, infrastructure as code (IaC) and immutable infrastructure patterns.
- Visualize full adoption. There are many variables: Available products will evolve, as will the AIOps “state of the art” and the infrastructure and applications for which you’re responsible. Consider the build-versus-buy continuum and how much of each to use.