Invited Talk 2
Assoc. Prof. Dr. Md. Yusuf Sarwar Uddin
University of Missouri, Kansas City, USA
Abstract: In the era of data, extracting values from data is of great importance where a massive amount of data is being collected, analyzed, and disseminated to end-users for wellness, safety, and other social good. This value proposition is in part due to the proliferation of heavily instrumented physical spaces with a rich and diversified set of sensors constituting the Internet of Things (IoT) or Cyber-physical Systems (CPS). One of the objectives of these data systems is to derive actionable information from raw data and deliver them to a large number of end-users, realizing the so-called Observe-Analyze-and-Act (OOA) paradigm. In this realm, the traditional approach of pushing *all* data to the cloud or to centralized data centers is neither viable nor economical. Instead, recent efforts suggested an edge-centric approach that brings computation closer to both data and users thus giving rise to an edge computing paradigm for data-driven systems. Arguably, the edge can manifest itself in two possible directions: (a) as data moves toward the cloud and gets processed en route to the cloud (data-facing edge) and (b) as data moves toward the end-users and gets processed on its way (user-facing edge). In this talk, I will discuss a couple of recent works in these two directions.