To prevent outages of business-critical digital services, organizations often invest heavily on IT Ops tools and staff. An AIOps platform can help organizations minimize that spend and boost service assurance, resulting in broad business benefits.
What drives an enterprise’s decision to adopt an AIOps platform? Which key problems are they trying to solve? Once it’s deployed, how can they track its value, and explain it to non-IT business leaders? What cultural and technology changes does an AIOps deployment trigger?
These were some of the thought-provoking questions addressed during the fireside chat “How AIOps Delivers Business Benefits” by Angeline MacIvor, Director of AI in IT Operations and an enterprise AIOps expert at Moogsoft, and Will Cappelli, Moogsoft’s EMEA CTO.
Framing their discussion is the reality that to prevent outages of business-critical digital services, many organizations invest heavily in IT monitoring and management tools and staff. However, as thousands of enterprises globally have found out, an AIOps platform helps minimize that spend and boost service assurance, triggering broad business benefits.
So how does an organization reap those significant benefits from AIOps? As MacIvor and Cappelli explained, there are a number of important elements to achieving AIOps success.
IT complexity hurts the business
The move to AIOps usually begins with an organization’s realization that its IT environment has become too complex, with too many tools generating an overwhelming amount of alerts. This results in IT instability, increased downtime risk, and potentially severe impacts on the business.
“For one of my pharma clients, if the manufacturing system of a critical drug that saves lives goes down, that batch has to be dumped, so revenue is not collected, and people don’t get better. They take that very seriously,” MacIvor said.
In the last two years, organizations have greatly improved how they assess the cost of downtime of critical systems, and realized that it is huge, often north of $300,000 per hour. “There’s been real change in the ability to measure the impact of an outage,” Cappelli said.
With IT driving so many parts of the business, this awareness about the importance of digital service assurance has driven the interest in AIOps and in its ability to streamline and automate the operations of highly complex, modern IT environments.
Take a snapshot of the state of your IT ops
It’s also key for organizations to take stock of their current IT operations — what tools they’re using, what resources they have, what’s their incident management workflow, who does what, how many monthly P1 tickets do they generate, and so on.
“Almost invariably they don’t have that picture, they don’t have that story,” said MacIvor, who in her five years at Moogsoft has helped many enterprises deploy AIOps successfully.
Once they do, they realize the high operational cost and inefficiencies, and are able to build a business case for deploying an AIOps platform. Once AIOps is in place, the benefits are so significant that it’s almost impossible to quantify them.
“I don’t know a single customer who’s come back and told me that we saved them ‘X’ amount of dollars, because the numbers are so large,” she said.
Examples of AIOps benefits from her customers include a reduction in alert “noise” of 99%, a 70% cut in server outages, and one full year with zero missed incidents.
To hear MacIvor and Cappelli share many more best practices, insights and real-world anecdotes, watch their fireside chat, which also included a Q&A session with the audience.
Here are the answers to several questions they weren’t able to answer live.
I haven’t heard much discussion around cloud adoption for IT workloads. What is your take on monitoring cloud workloads that will usually run on immutable operating systems?
Excellent point. It is indeed something we should have discussed — maybe next time! In any case, AIOps is a particularly critical function in a cloud-based environment because of the speed with which servers, storage, applications, and so on, can be added and removed. Problems become extremely difficult to diagnose because their causes are no longer there to be examined. While one cannot directly manipulate the operating systems in response to an issue, the AIOps platform will provide direction on how to reconfigure the service being consumed.
What is the functionality in the Moogsoft AIOps Platform that explains or displays the summary of the Situations or events? And how do you make this functionality better based on the organizational requirement?
The composition and structure of events or Situations are displayed on the UI of the Situation Room. There are, in fact, a number of distinct views suitable to different requirements above and beyond a simple listing of the elements that constitute a Situation. One view is a timeline that shows how data items indicating events accumulate into a Situation over time. Another view shows the various data items indicating events organized by topology. Finally, one can amplify that topological view with different types of causal information.
How will AIOps evolve in the future, and how is Moogsoft going to keep up with the market?
AIOps will evolve in a number of ways. It will become more distributed in the sense that AIOps platforms will likely take the form of multiple intelligent agents that, in addition to applying their own algorithms to data, will generate results by interacting with one another.
Automation and robotics will also become central to AIOps as enterprises increasingly look to close the loop from observation through analysis to action. Finally, AIOps will extend its domain beyond IT operations and service management proper to other areas like DevOps and SecOps.
Moogsoft, having led the market so far, with regard to the vision of what AIOps is all about, has a very clear picture of where the market is going and our academic and product development oriented research continues to generate the intellectual property that will get us there.