The visual fog paradigm envisions tens of thousands of heterogeneous, camera-enabled edge devices distributed across the Internet, providing live sensing for a myriad of different visual processing applications. The scale, computational demands, and bandwidth needed for visual computing pipelines necessitates offloading intelligently to distributed computing infrastructure, including the cloud, Internet gateway devices, and the edge devices themselves.
This session deep-dives into two aspects of visual fog orchestration: offloading and scheduling. Offloading is a mechanism for realizing (live) workload migration, whereas scheduling attempts to resolve the problem of assigning the visual computing tasks to various devices to optimize network utilization. You’ll be introduced to our pioneered study, in which we demonstrate sub-minute computation time to optimally schedule 20,000 tasks across over 7,000 devices, and just 7-minute execution time to place 60,000 tasks across 20,000 devices. Witness an approach that’s ready to meet the scale challenges: visual fog is feasible and a viable paradigm to scale out video analytics systems.
Senior Staff Research Scientist ,