As an idea conceived by Gartner four years ago, AIOps is already a mature practice. But it is also one that continues to evolve as businesses turn to AIOps to support new use cases, and as AIOps vendors build better and more efficient AIOps tools.
That fact begs the questions: what’s next for AIOps? What are the relevant trends that will shape the future of AIOps over the next several years, and how will AIOps use cases evolve going forward?
Predicting the future is always a messy business, of course, but here are some predictions about what’s next for AIOps.
Growth of Domain-Agnostic AIOps Tools
Traditionally, most AIOps solutions have fallen into the category that Gartner labels “domain-centric.” These are tools that can support a limited range of use cases, such as those associated with a certain industry or a certain type of data. Domain-centric AIOps tools are easier to build because they have a relatively narrow range of functionality. Of course, that is also a key drawback. It requires organizations to deploy multiple domain-centric AIOps solutions if they need to address use cases that span multiple domains.
Going forward, an increasing number of AIOps tools are likely to evolve into (or be replaced by) domain-agnostic solutions. Domain-agnostic AIOps tools (like Broadcom’s) can cater to a broad array of use cases, and they can leverage virtually any type of data source without the need for special training or configuration.
Enabling DevOps Risk Management
In the world of DevOps and SRE, risk management is a relatively novel concept. It focuses on how DevOps teams and SREs can use the software delivery chain to manage organizational risks.
Traditionally, this process required a fair amount of manual work. Teams relied on practices like manual post-mortems and qualitative feedback loops to figure out where their greatest risks lay with regard to software delivery.
With AIOps, however, DevOps risk management can become a highly automated, efficient process. AIOps tools can automatically assess the relationship between software delivery processes and user experiences, then identify weak points that teams should address in order to optimize software delivery outcomes.
More generally, AIOps will help DevOps engineers and SREs to manage change more effectively as well as respond to outages more rapidly. AIOps is already doing this, of course, but as AIOps tools grow even more sophisticated, they will enable greater and greater degrees of reliability within the software delivery process.
True DevSecOps: Finally Integrating SecOps and IT Ops
DevSecOps, a concept that encourages security operations teams to work closely with IT operations and development teams, has been a buzzword for years.
But the problem with DevSecOps is that it’s easy to talk about and hard to practice. How do you actually give SecOps teams the visibility they need into DevOps processes to enable true collaboration between all stakeholders? How do you ensure that security problems can be surfaced and resolved quickly without disrupting the software delivery process? Those are not simple questions to answer.
With AIOps, however, making DevSecOps work in practice becomes a more realistic proposition. There are two main reasons why. First, AIOps helps teams collect and analyze data related to all types of risks – including those related to security and application performance – and then consolidate it into alerts and recommendations that everyone in the IT organization can use. This is one practical way to de-silo DevOps and SecOps, because it means that both teams have a common set of dashboards to work from when solving problems.
Second, AIOps can play a critical role in helping teams to understand the most efficient and secure way to solve problems. That’s important because, traditionally, the most obvious solution for a software performance issue wasn’t always the most secure solution, and the fastest solution to security problems wasn’t necessarily the least disruptive to end-users. AIOps can help teams sort through complex issues in order to find the right middle ground between security priorities and software delivery goals – which is exactly what DevSecOps is all about.
A Step Closer to NoOps
For years, there has been talk of NoOps, a concept that describes a world in which IT operations are so completely automated that they no longer really exist.
We may never reach a state where humans cease to play any kind of role in IT operations. However, AIOps is very much ushering in a world in which infrastructure becomes largely self-managed, greatly reducing the burden placed on IT teams to solve issues manually. Truly complex problems may always require human intervention, but routine operations, like autoscaling and the remediation of standard performance problems, can be outsourced to AIOps tools.
AIOps already delivers a great deal of value to IT organizations seeking faster, more efficient, and more reliable means of managing complex IT estates. Yet AIOps is poised to do even more in coming years as AIOps solutions grow more sophisticated and better able to address the needs that extend beyond basic IT operations.