Applying AI and ML to ITOM
The latest Wikibon Weekly Research Meeting covers the benefits of applying AI and machine learning to the discipline of IT operations management (ITOM). Before jumping on the latest trend, we suggest three things that CIOs should consider before jumping into the hype.
This post summarizes three CIO actions before turning to AI for ITOM. Let’s review the operations management problems IT organizations face today and the business impact that improvements will have. IT is viewed as reactive, affecting a laundry list of business issues, including:
- Application availability
- IT and business productivity
- Data quality
- Application performance
These issues increase business risk and cost. Practitioners in the Wikibon community express a desire to be proactive to address these problems. Meanwhile, vendors are promising that their tools will allow them to be more anticipatory. Customers want reduced false positives and to limit the number of trivial events they must chase. And of course cloud complicates all this.
The vendor community promises end to end visibility on infrastructure platforms including clouds. That commitment includes the ability to discover/manage events and identify anomalies in a proactive manner. Maybe even automate necessary remediation steps– all good and important features of ITOM software. Furthermore, suppliers put forth a vision that these capabilities must and can align with and map to critical business processes. This is important so that customers can set priorities based on business value. And of course all this must extend to cloud resources.
Is AI and ML the Answer?
Wikibon is encouraged that bringing analytics to ITOM disciplines has great potential. But before diving into an industry buzzsaw, we suggest that organizations look at three things before making the leap:
- Seventy percent of infrastructure problems come from poorly applied or failed changes. So start with good change management and you’ll attack 70% of the problem;
- Second, we advise that CIOs should narrow down their promises and get their SLA’s firmed up so they can meet and exceed expectations. Build credibility with an organization before taking on wider responsibilities. Don’t take on too much because in our experience organizations will inevitably hit a skills gap. We recommend finding a sweet spot of services and not spreading too thin.
- Third is to start acting like a cloud provider. Be clear about the services that you offer. Clearly communicate the SLA’s agreed in #2 and appropriately charge for those services so that you can fund your operations.
Execute on these three imperatives from a process and skills standpoint before you start throwing technology at the problem and you’ll deliver much faster ROI with less risk and technical debt.
The clip below excerpts a summary from the research meeting.
