AIOps (AI for IT operations) adoption is on the rise as organizations invest in AI to make their IT ops smarter, faster, and more secure. Those who have adopted AIOps view the technology as no longer a nice-to-have but a necessity in the post-pandemic, work-from-home era. IT leaders are tasked with managing third-party cloud applications from devices and remote workers scattered across numerous locations in this new era.
The insights come from a recently published State of AIOps Study, conducted by ZK Research, sponsored by Masergy, a software-defined networking (SD-WAN) services company. In August 2021, ZK Research surveyed more than 500 IT decision-makers in the U.S. across seven industries.
IT decision-makers believe AIOps offers their organization several business benefits, including improved productivity, cloud application performance, and security. The majority (94%) of the respondents believe it's important or very important for AIOps to manage network and cloud-app performance.
AIOps continuously monitors app performance using ML algorithms by assessing bandwidth usage patterns, identifying anomalies, and predicting outages. Masergy, for example, has embedded AIOps directly in the application layer of its newly enhanced AIOps platform to help IT teams improve cloud app performance on global networks.
AIOps adoption is starting to reach the masses, with network and security automation as the key drivers. One of the more interesting findings is that 64% of organizations claim to be already using AIOps. This number first seemed high to me, but I believe that's because a number of organizations are using tools that claim to be AIOps but are simply rules-based engines.
It's important to understand that AIOps is not an upgraded management console or fancy SIEM (Security Information and Event Management). It's a data-driven application or service that goes through a training process and then is restrained over time. It's critical that buyers do their due diligence and ensure they are using a true AIOps tool and not one that's just branded AI.
Another interesting data point is that 55% of organizations are using AIOps across both network and security, meaning more than half of companies want to bring security and networking together. This is a trend the industry has been watching for a decade or more, and it hasn't come to fruition yet. Now that digital businesses are network-centric securing an organization should be done at the network level. It would have been nice to see this number higher, but 55% is certainly moving the industry in the right direction.
Also, 84% see AIOps as the path to a fully automated network environment, and 86% expect to have a completely automated network within five years. It's good to see respondents understand that today AIOps can't deliver a fully self-driving network, but it's a step along the way.
A good analogy is to think of AI in automobiles. Today, there are loads of AI features, such as lane change alert and autopilot, that make driving safer, but a fully autonomous vehicle remains a vision. IT pros need to think of AIOps as a tool that makes their jobs easier today, but they still should plan for automation.
The best way for organizations to fully benefit from AIOps is to invest in SD-WAN and secure access service edge (SASE), which combines elements of SD-WAN and network security into a single cloud-based service. When organizations adopt SD-WAN and ultimately SASE, they gain virtualization and orchestration capabilities that are necessary for managing distributed networks and security policies. Also, SD-WAN and SASE combine network and security data, making it easier to coordinate security changes with network updates.
Legacy networks aren't designed for AI or centralized intelligence. AIOps is only as smart as the data it is fed, so it can't become fully automated when used on a network that's not managed using the software. SASE provides a single converged platform where AIOps can access data across both the network and security to automate IT processes. Most (77%) organizations agree that AIOps performs better with a SASE architecture, in which SD-WAN and security are combined in one platform.
More than two-thirds (73%) of the study participants believe IT should be investing in SD-WAN modernization and virtualization tools in preparation for AIOps. The other top two investments required to become AI-ready are cloud migration (67%) and training for AI models (61%).
However, AIOps migration involves more than just deploying tools. Half of the organizations cite eliminating system fragmentation as a major AIOps challenge. AI shouldn't be fragmented across multiple tools, but rather AIOps and SASE should work together to provide overarching insights.
The good news is organizations trust AIOps tools to create fully automated systems. The majority (97%) of the respondents are confident in these tools. For 65% of organizations buying and implementing solutions, AIOps features drive their selection. Meanwhile, 50% deploy AIOps using a fully managed or co-managed service.
As is the case with any type of automation, AIOps requires reframing of processes, roles, and responsibilities. Organizations should create their own IT operations playbooks to train an AIOps engine. When choosing a solution, ZK Research recommends asking vendors -- preferably not incumbents -- for AI efficacy rates and how those are increasing over time. Every organization should have an AIOps strategy to avoid being left behind.