Design & Architecture

AIOps the next big thing for Enterprise IT?

Tony Mattke · 2018.10.18 · 3 min read

But how does all of this relate to IT Infrastructure and Operations? AIOps is the combination of AI and IT Operations. And while the IT community hears the AI related buzzwords, not many of us know where or how it can help. Lucky for us other large IT focused businesses continue to pour big money into AI. Which means the capabilities of the systems we see at an enterprise level will only continue to improve. As the capabilities of these AI systems grow, so does the impact that it can make. And while that sounds great, what can do we do with it?

With the complex reality IT Ops folks live in day-to-day, and the constant struggle to do more with less, it’s a tough job to stay on top. We’re drowning in the wave of data that we’re generating. I’m not just referencing syslog, or SNMP, but now we have an ever-growing number of interconnected IoT / smart devices. All of this data can be leveraged to build assurance in how our systems are operating, trending, and even analyzing behavior.

I’m guessing many of us are already using an application with machine learning or big data functionality in our I&O playbooks. Splunk for instance is a machine data and analysis software provider. We’re already funneling immense amounts of data into Splunk, and while many of us have dashboards, alerts, and other tools leveraging that data; very few of us are using that to predict failures. Being able to predict a failure 10-20 minutes before it happens seems like something out of a sci-fi movie, but it’s completely feasible.

Another area where I think I&O will receive augmented help from AI is in capacity planning. Letting AI monitor the trends we’re seeing within our infrastructure is an easy win. And with hybrid cloud environments becoming the norm, leveraging AI to do the mapping of workloads to the right infrastructure is a no brainier. In these hybrid cloud environments we see the cloud bringing us a certain amount of elasticity to our infrastructure. Many companies depend on IaaS auto-scaling in order to take advantage of that flexibility. But now, if we leverage AI to add a level of prediction, based on current trends and historical data. That is a leap forward in ensuring scalable application delivery.

By now you get the idea – give AIOps a large data set, and let it crunch. From scaling and auto-tiering to anomaly and threat detection / analysis. Leveraging AI as a member of the team allows us to monitor more aspects of our infrastructure than ever before. Living in the enterprise world of SD-everything, leveraging AI brings us the next level of being able to expand our business footprint without major expansions in staffing. But with these new technologies adding key skill sets and training will be more important than ever.

More in Design & Architecture
comments powered by Disqus

Related Posts

Fundamentals

Well-Known Intervals

Listed below are many events which occur on network devices at well-known intervals. The list is provided to serve as an aid while troubleshooting recurring network disruptions.

2015.08.10 · 4 min
Industry & Events

CCIE Potential

INE published a great info-graphic on the earning potential of Cisco’s certifications and I felt the need to share it here.

2012.04.24 · 1 min
Industry & Events

Brocade and VCS… quite impressive

Our second visit on day 2 of Network Field day was Brocade, who incidentally supplied us with a great lunch!

2011.11.16 · 4 min