Artificial intelligence was once a magical concept, the stuff of science fiction. Today, after decades of research and commercialization, this is just another fundamental tool in keeping the business stack running smoothly.
Nowhere is this more evident than in the world of DevOps, a data-rich back office practice that presents a perfect sandbox for exploring the power of artificial intelligence. Operations teams now have a thriving collection of labor-saving and efficiency-enhancing tools and platforms under the acronym AIops, all of which promise to apply the best intelligence algorithms. artificial at work of IT infrastructure maintenance.
AIops is among the best use cases for artificial intelligence. Servers and networks generate petabytes upon petabytes of data. We know when processes start and stop, increase and decrease, often down to the millisecond. The demands for RAM and CPU are often well understood, as are the prices for renting hardware in the cloud. All are often calculated up to six or seven significant digits. Creating an autonomous car may mean struggling with a world filled with pedestrians, cattle and shadows, but when it comes to IT infrastructure, everything is already digitized and ready for analysis.
Some of the simpler tasks for AIops are to speed up the way software is deployed to cloud instances. All the work done by DevOps teams can be improved with smarter automation that can monitor loads, forecast demand, and even start new instances when the hordes descend.
The right AIops tools make forward-looking guesses about the machine load, then watch to see if anything deviates from those estimates. Anomalies can be turned into alerts that generate emails, Slack posts, or, if the gap is large enough, pager messages. A good chunk of the AIops stack is devoted to managing alerts and ensuring that only the most important issues turn into something that interrupts a meeting or a good night’s sleep.