AIOps: Hype vs. Reality

What is AIOps? How does an AIOps platform help your observability practice?

"AIOps platforms analyze telemetry and events, and identify meaningful patterns that provide insights to support proactive responses. AIOps platforms have five characteristics:"[1]

1. Cross-domain data ingestion and analytics

2. Topology assembly from implicit and explicit sources of asset relationship and dependency

3. Correlation between related or redundant events associated with an incident

4. Pattern recognition to detect incidents, their leading indicators or probable root cause

5. Association of probable remediation

The above is Gartner’s definition and is part of the Gartner® “Market Guide for AIOps Platforms.” The Gartner definition is also aligned with our view. We have compiled a few of the key takeaways and recommendations from the Gartner report in this blog, as well as added some of our own perspective.

At StackState, we’ve been helping our customers implement AIOps within their observability practices for several years. However, for many companies the insights and opportunities for more informed actions that AIOps unlocks are just coming to light. In a recent survey we did on observability, only 11% of survey respondents said they had adopted an AIOps-driven approach. Implementation of it is clearly still in its early stages, but one thing’s for certain: AIOps provides value to organizations and, as they mature in their observability practices, the use of it will grow.

According to a published Forbes article, AIOps solutions are “expected to be worth more than $40 billion [in market size] by 2026,” a significant jump from the $13.51 billion recorded as of 2021. In other words, it is not just evolving, it’s exploding. 

More and more businesses are embracing AIOps platforms, with the media writing about it  almost every day. Last year, IBM published “Three Reasons AIOps Is the Future of ITOps,” calling it out for providing complete visibility, improving IT service management and reducing noise through AI automation. IBM isn’t alone in celebrating the benefits of it. Companies like PagerDuty have also posted content labeling AIOps “the future of DevOps.” We disagree that it is only the future of DevOps. We believe it is also the future of observability. And, at StackState, the future is now.

All of the industry coverage about AIOps is shedding light on the growing potential for the long term. But where is the market today? The Gartner report will inform and get you up to speed on the latest trends.

As you might expect from the report’s title, we believe Gartner does a great job at describing the market. But the report also explores topics as diverse as market direction, analysis and provides recommendations. 

We are sharing four key takeaways from the Gartner report in this blog. They are: 

1. Barriers to implementing AIOps platforms

From the report: “One of the main barriers to implementing artificial intelligence for IT operations platforms is the difficulty measuring their value and lack of understanding of benefits derived.”[2]

Gartner’s recommendation for I&O leaders: “Focus on tangible and incremental business outcomes with quantitative value-based proof points. Avoid the AIOps hype.”[3]

The StackState view: In our experience, our customers who focus on business goals for AIOps, first, are more successful. Examples of business goals could be things like reducing false alerts, reducing the amount of tickets created in ITSM or detecting patterns in anomalies before they become an actual incident, have a faster implementation and a more successful implementation than making AIOps the goal in and of itself. Business goals should always be leading and driving decision making. The minute you adopt a new technology just to adopt it, you’re in trouble.

2. Data management costs and complexity concerns

From the report: “Data management costs and complexity are becoming a concern for many enterprises that have adopted AIOps platforms as they expand their use. The hurdles include optimizing data storage, controlling cost, improving data quality at the source and, as of late, data transmission.” [4]

Gartner’s recommendation for I&O leaders: “Target productivity outcomes when using AIOps platforms; for example, leverage them to enhance workflows and to improve efficiency of IT personnel.”[5]

The StackState view: AIOps should be used to enhance human judgment. With data still existing within individual team’s monitoring or observability tools this is challenging, because the data is in siloes and therefore not usable when looking across the entire IT stack. Teams cannot get a holistic view of what’s going on or the health of the overall environment. At StackState, we believe that topology is the anchor to unify, correlate and visualize all data into a single data fabric. Topology builds structure in how components are related and correlates metrics with it. This superstructure of data can quickly provide the answer to practically any question with regards to your system’s health and what is going on within the entire stack. A structured data set visualized within a single topology significantly lowers complexity, increases the ability of the data to be used and enables it to contribute value in understanding the health of the environment.

3. AIOps replacing traditional monitoring tool categories

From the report: “Enterprises are replacing some traditional monitoring tool categories by embedding them within AIOps platforms. For example, virtual network monitoring, observability and infrastructure as a service (IaaS) monitoring are being done entirely within AIOps platforms, especially if the enterprise has its entire IT footprint in the cloud.”[6]

Gartner’s recommendation for I&O leaders: “Leverage AIOps platforms for scenarios like adaptive anomaly detection or system-centric anomaly detection. Entity-centric anomaly detection is better served by monitoring tools.”[7]

The StackState view: Although some technology or practices might become obsolete, often they can still fill a need. For example, foundational monitoring or collection of metrics will remain vital despite available new technologies. New technologies will just provide new and additional capabilities that extend the value of some older technologies. Anomaly detection still requires metric streams, root cause analysis (RCA) still requires a broad set of data to determine the cause. Often newer AIOps solutions augment existing investments.

4. Increasing use of AIOps platforms across IT operations management (ITOM)

From the report: “Enterprises continue to increase their use of AIOps platforms across various aspects of IT operations management and mature their use cases across DevOps and site reliability engineering (SRE) practices.”[8]

Gartner’s recommendation for I&O leaders: “Create an operations model to provide metadata and insights as a service to different departments such as finance, sales and marketing.”[9]

The StackState view: Implementation often starts in units and departments that are part of IT or close to IT. More system data is available to be used for decision-making, learning and optimization of AIOps initiatives. That’s a great place to start. However, the next step is to extend AIOps data to other areas such as business or customer-oriented functions, and use the intelligence to enhance insights and decision-making in these domains, beyond IT.

With nearly 20 pages of insights, the Gartner® report provides a lot of value on the state of AIOps today. Whether you’re a novice or an expert, if you want to know what the market looks like right now or where it’s headed in the future, this Gartner report helps you make sense of the landscape. 

Want to read more about data ingestion and handling, or machine learning analytics? What about how access to automated insights is changing the future of AIOps? It’s all there. What’s more, Gartner provides a list of Representative Vendors in the AIOps platforms market (yes, we’re on the list!). 

AIOps is a promise that is becoming reality. It’s here to stay and it will help us all to continuously improve the reliability of our IT environments. At StackState, we believe many organizations are just starting this journey. StackState takes an entirely different approach to observability by putting topology at the center of it. Topology matters because it provides the structure that is critical for insights, intelligent decision-making and AIOps capabilities.

Attribution and disclaimers

[1], [2], [3], [4], [5], [6], [7], [8], [9] Gartner, “Gartner Report: Market Guide for AIOps Platforms,” Pankaj Prasad, Padraig Byrne, Gregg Siegfried, Published 30 May 2022.

Gartner is a registered trademark of Gartner, Inc. and/or its affiliates in the U.S. and internationally and is used herein with permission. All rights reserved.

Gartner does not endorse any vendor, product or service depicted in its research publications, and does not advise technology users to select only those vendors with the highest ratings or other designation. Gartner research publications consist of the opinions of Gartner’s research organization and should not be construed as statements of fact. Gartner disclaims all warranties, expressed or implied, with respect to this research, including any warranties of merchantability or fitness for a particular purpose.