Abstract
For many years now, my students and I have developed several solutions to enable cost-effective deployments in the cloud of low-latency highly-available services. Based on this work, in this talk, my goals are two-fold.

First, I will give an overview of how some of my group's work, as well as other related research from computer science researchers, can be leveraged to enable cloud-driven control of manufacturing devices, i.e., how CS research can be used for improved manufacturing.

Second, in the context of smart manufacturing, I will describe the challenges in applying the approaches traditionally used by computer science researchers. In particular, I will argue that many of the principles that underlie use of the cloud will need to be rethought, and the domain of smart manufacturing presents an exciting new frontier for distributed systems research. I will then show some preliminary work done in collaboration with Prof. Okwudire on the application of cloud computing to the control of manufacturing machines.

Bio
Harsha V. Madhyastha is an Associate Professor in Computer Science and Engineering at the University of Michigan. His research interests span all facets of scalable, fault-tolerant, performant, and privacy-preserving networked systems. His work has resulted in award papers at several top-tier conferences and also received the IRTF's Applied Networking Research Prize on multiple occasions. He is the recipient of Google Faculty Research awards, a NetApp Faculty Fellowship, a Facebook Faculty Award, and an NSF CAREER award.