RAVAS: Interference-Aware Model Selection and Resource Allocation for Live Edge Video Analytics
Ali Asghar Rahmanian, Ahmed Ali-Eldin, Selome Kostentinos Tesfatsion, Björn Skubic, Harald Gustafsson, Prashant J. Shenoy, Erik Elmroth · 2023
Numerous edge applications that rely on video analytics demand precise, low-latency processing of multiple video streams from cameras. When these cameras are mobile, such as when mounted on a car or a robot, the processing load on the shared edge GPU can vary considerably. Provisioning the edge with GPUs for the worst-case load can be expensive and, for many applications, not feasible.