Low-cost cameras grow rapidly, producing colossal videos that enable powerful analytics but also stress network and compute resources. An unexploited opportunity is that most of the videos remain "cold" without ever being queried. For resource efficiency, we advocate for these cameras to be zerostreaming: they capture videos directly to their cheap local storage and only communicate with the cloud when analytics is requested.To this end, we present a system that spans the cloud and cameras. Our key ideas are twofold. When capturing video frames, a camera learns accurate knowledge on a sparse sample of frames, rather than learning inaccurate knowledge on all frames; in executing one query, a camera processes frames in multiple passes with multiple operators trained and picked by the cloud during the query, rather than one pass processing with operator(s) decided ahead of the query. On diverse queries over 15 videos and with typical wireless network bandwidth and low-cost camera hardware, our system prototype runs at more than 100× video realtime. It outperforms competitive alternative designs by at least 4× and up to two orders of magnitude.

We present VStore, a data store for supporting fast, resourceefficient analytics over large archival videos. VStore manages video ingestion, storage, retrieval, and consumption. It controls video formats along the video data path. It is challenged by i) the huge combinatorial space of video format knobs; ii) the complex impacts of these knobs and their high profiling cost; iii) optimizing for multiple resource types. It explores an idea called backward derivation of configuration: in the opposite direction along the video data path, VStore passes the video quantity and quality expected by analytics backward to retrieval, to storage, and to ingestion. In this process, VStore derives an optimal set of video formats, optimizing for different resources in a progressive manner.VStore automatically derives large, complex configurations consisting of more than one hundred knobs over tens of video formats. In response to queries, VStore selects video formats catering to the executed operators and the target accuracy. It streams video data from disks through decoder to operators. It runs queries as fast as 362× of video realtime.

No abstract

No abstract

No abstract