Live video analytics at scale with approximation and delay-tolerance

TLDR

Video cameras are widely deployed for security and smart‑city applications, yet realizing their full potential requires efficient real‑time analysis and careful resource management due to high vision‑processing costs. The authors present VideoStorm, a system that processes thousands of video‑analytics queries on live streams across large clusters. VideoStorm builds query resource‑quality profiles offline and uses an online scheduler that allocates resources to maximize quality and lag performance, exploiting resource‑quality tradeoffs and diverse quality/lag goals instead of fair sharing. On a 101‑machine Azure cluster, VideoStorm improved query quality by up to 80 % and reduced lag sevenfold while processing traffic‑camera video.

Abstract

Video cameras are pervasively deployed for security and smart city scenarios, with millions of them in large cities worldwide. Achieving the potential of these cameras requires efficiently analyzing the live videos in real-time. We describe VideoStorm, a video analytics system that processes thousands of video analytics queries on live video streams over large clusters. Given the high costs of vision processing, resource management is crucial. We consider two key characteristics of video analytics: resource-quality tradeoff with multi-dimensional configurations, and variety in quality and lag goals. VideoStorm's offline profiler generates query resource-quality profile, while its online scheduler allocates resources to queries to maximize performance on quality and lag, in contrast to the commonly used fair sharing of resources in clusters. Deployment on an Azure cluster of 101 machines shows improvement by as much as 80% in quality of real-world queries and 7× better lag, processing video from operational traffic cameras.