Recently, the concept of using artificial intelligence for IT operations (AIOps), which Gartner defines as the combination of AI, big data and machine learning to manage the main functions of IT operations, “including monitoring availability and performance, correlating and analyzing events, as well as managing and automating IT services, ”has become necessary. The Gartner Institute has predicted that large companies’ exclusive use of AIOps and digital experience control tools to monitor applications and infrastructure will increase from 5% in 2018 to 30% in 2023.
In recent months, the dispersal of company teams with the generalization of teleworking has undoubtedly accelerated the awareness that IT must work as much as possible “in the dark” (automation). But the implications of AIOps go far beyond the recent health crisis. Much of the IT activity has gone into developing, scaling, and supporting AI in business, and we’ve come to a point where AI will help us build, deploy, and manage our next generation of AI.
How can AIOps help solve IT problems? Jessica Rockwood, vice president of engineering at IBM Watson, covered the main strengths of AIOps in a post, while IBM recently announced the launch of its Watson AIOps tool. Overview of the main assets:
- “Collect data from a heterogeneous set of sources in the IT infrastructure, from performance alerts to incident tickets. This data can be used to reduce costs and help improve productivity by recognizing a specific time of day when demand for IT resources is low, and automatically moving IT resources. ”
- “If automatic adjustments are not desired, the data can be displayed in a visual format that provides IT operations managers or site reliability engineers with recommended guidelines, and explains the rationale behind these recommendations” .
- “Automate tasks such as moving traffic from one router to another, freeing up space on a disk, or restarting an application.”
- “Artificial intelligence systems can also be trained in self-correction so that IT managers and their teams can devote their time to tasks with higher added value, while enjoying full visibility into operations. company “.
What does an AIOps platform contain? Sameer Padhye, Bishnu Nayak and Enzo Signore explore essential basics in their ebook AIOps for Dummies [“AIOps pour les Nuls”, NDLR] :
- Ingestion of open data : An AIOps platform collects data that can include “operational information such as outages, performance measures, tickets, and more.” The ability to ingest data from a wide variety of sources is essential because it provides an accurate, real-time view of all moving parts in hybrid IT environments. ”
- Self-discovery : Businesses need an auto-discovery process that automatically collects data across all infrastructure and application areas, including on-premise, virtualized and cloud deployments. Auto-discovery also identifies all infrastructure devices, running applications, and the resulting business transactions.
- Correlation : “The AIOps platform correlates this data in a contextual form. It must determine the relationships between infrastructure elements, between an application and its infrastructure, and between business transactions and applications. ”
- Visualization : Visualization allows IT operations to “spot problems quickly and take corrective action.” “
- Machine learning : “The AIOps solutions use supervised and unsupervised machine learning to determine the patterns of events in a time series. They also detect anomalies with regard to expected behaviors and thresholds and predict breakdowns and performance problems ”.
- Automating : Automation provides a return on investment “by automating human IT operations, reducing significant operating expenses and accelerating innovation.” It also reduces MTTR [temps moyen pour réparer, NDLR] and can improve customer satisfaction. “
The health crisis has put pressure on IT managers to cut costs and find ways to do a lot more with a lot less. At the same time, there is an insatiable demand for approaches based on artificial intelligence which will inevitably tax IT infrastructures.
For example, last year, Gartner analysts found that general AI projects were increasing. The average number of AI projects in place was four, but respondents expected to add six more projects in the next 12 months, and another 15 in the next three years. Ironically, AI itself provides a way to support a viable infrastructure to support the growing volume of AI initiatives.