AIOps lead by DevOps

 DevOps because of agile development practices bought in E2E responsibilities toward software development and delivery.  A Common DevOps cycle of the four stages of observe, detect, engage, and act has introduced ease to the operations team [1]. Still, that does not seem to be enough as post-release of software the operations done manually struggle to scale, are hard to standardize and are error-prone. Given these challenges, fully-automated operations pipelines powered by AI capabilities is a promising approach to achieving the SLA and SLO goals. AIOps the acronym for AI for IT Operations, was coined by Gartner in 2016. Gartner Glossary says, ”AIOps combines big data and machine learning to automate IT operations processes, including event correlation, anomaly detection, and causality determination”. To achieve that level of IT automation, investments in AIOps technologies are imperative. Hence it started appearing in the Gartner Hype Cycle for AI in 2017, along with other popular topics such as deep reinforcement learning, nature-language generation, and artificial general intelligence. Since 2022, enterprise AIOps solutions have been adopted by many companies for their IT infrastructure. This transformation from manual to automated operations using AIOps could be broken down into levels based on AIOps maturity as shown below.

Different systems and different industries may be at different levels of AIOps maturity, and their priorities and goals differ with regard to specific capabilities for adoption. Authors further foresee the trends to apply more complex AI techniques to address challenging business problems. Enabling such a community of AIOps practitioners is the need of the hour.

Interested analysts and developers can further read about the vision, challenges, and how to address them given [2].

Reference

1. https://www.dynatrace.com/news/blog/what-is-devops/
2. https://doi.org/10.48550/arXiv.2304.04661