That’s me, debugging assembler language and burning the compiled code into ROMs. Compiling the code alone seemed like it took forever, the process of burning the ROM and then testing took even longer. Yes, longer than forever.
In those days, the idea that artificial intelligence (AI) would be able to have a conversation with a user or fix an issue without human help was solidly in the realm of science fiction. This futuristic fairy tale was and is the most interesting area of computer science, at least as far as I’m concerned.
And while most fairy tales never come true, this one did.
Bringing proactive IT to life
Today, we’re light years away from the days I spent working in advanced R&D, slowly burning code into ROMs. We have AI to thank for the ability to deliver support with limited or no human interaction. The next fairy tale is no human interaction at all—and with AIOps on the scene, I expect this fairy tale to come true as well.
AIOps is the ability to apply machine learning to IT operations and big data. Using AIOps and data-driven workflows, we can automatically detect, analyze, and remediate issues before or just after they occur. This is what brings proactive IT to life.
At ServiceNow, machine learning and the intelligence in the Now Platform make it possible for an employee to submit an issue directly to a virtual agent where it can be resolved sans human engagement. This minimizes the number of issues coming from employees. AIOps also allows us to proactively—even predictively—identify issues so they can be fixed before employees know there was a problem. That is the epitome of proactive IT support.
Machine learning also accelerates resolution by assigning the issue to the most qualified operational team—again bypassing the human agent. Put together, this leads to lower costs and increased productivity.
None of this can happen without a single system of record for IT where AIOps can access in real time all historical incident, problem, and change data to determine what previous incident, problem, change, and resolutions have occurred. Unfortunately, too many IT departments grapple with siloed IT tools and data spread across multiple systems, keeping them stuck in a reactive, manual mode that drives costs up and productivity down.
Understanding real-time impact
Machine learning also helps determine the impact of an issue in real time.
It does this by correlating many different data points, such as persona, time, location, service, and application, to better understand the impact of an incident. For example, if finance complains that an ERP system is down, machine learning can automatically ensure this becomes a top priority. This approach is much more effective than depending on employee input about the impact, which can be subjective.
AIOps is critical to reducing the thousands of events (aka monitoring noise) coming in from across the IT estate, both on-prem and in the cloud. When AIOps is applied, monitoring noise can be cut back by 99%. AIOps uses event correlation, pattern recognition, and anomaly detection to present only the critical few alerts that need to be addressed.
The ultimate outcome of AIOps—and, indeed, one of the greatest values it gives your IT operations team—is the ability to understand the exact impact of an infrastructure-related issue on a critical service, application, or end user.
Zero impact, zero touch
The end goal of proactive IT support is to both predict and prevent issues, using AI to identify anomalies and proactively take a fully automated action. The ideal scenario is one where there is zero impact on end users and zero touch by the ops teams.
As an example, at ServiceNow was one of our most complex problems was also one of the first we could resolve proactively—our VPN service. Today, with most of our employees working remotely, VPN is a mission-critical service. By identifying abnormalities in our firewalls and active directories, then correlating that data with endpoint device data, we can automate the restoration of VPN services. By doing so, we bring operational costs down and employees productivity up.