Lecture: May 16, 2023

web.stanford.edu/class/ee392b


AI for IT Operations - AI Ops in the Cloud

Allison Jones, Principal Group Program Manager, Microsoft

Bio

Allison Jones is currently Senior Director of Product Management, AIOps Azure Engineering at Microsoft. Allison has been with Microsoft for nearly twenty years and has been involved with AI/ML initiatives for improving Azure Engineering Operations, transforming the formerly alert based anomaly response systems to the automated detection and actions platforms using cutting edge ML anomaly detection algorithms to improve livesite response for all of Azure operations. Allison has also worked in several other divisions at Microsoft including eCommerce micro-services group and strategic OEM/IHV engineering partner architectures. Prior to joining Microsoft, Allison was at Intel Online Services for three years building the industry’s first managed hosting environments by leveraging Microsoft Windows Service and Intel processors. She’s also worked for several other industry leading companies such as Adobe and Corbis.

Abstract

Allison will discuss how Microsoft uses AIOps to Power Cloud Scale Operations. Microsoft’s Azure Cloud spans over 160 Datacenters across more than 65 regions worldwide and powers some of the world most critical workloads. With billions of time series processed every minute and 100s of PB of logs per day, managing Azure’s Service Operations at Cloud Scale needs intelligent systems to identify and mitigate critical issues in seconds. Learn how Microsoft is leveraging AI/ML to manage its world class service operations systems.

Lecture Notes

Presentation slides (pdf)

Video recording of the lecture