AVP Record is a highly optimised, hierarchical store of media and associated metadata. It uses open standard formats (fMP4 - you can open its files in any compatible media application) alongside proprietary indexing for exceptional performance.
Be that a rolling one-hour “catch-up” or DVR recording, live to vod highlight editing or 90 days recording for compliance monitoring.
The efficient indexing provided by AVP Record enables highly flexible access to the stored media and metadata.
This enables just in time muxing strategies, catch-up TV and live to vod workflows where the final edited assets are available seconds after the live stream competes.
As well as “just” being an indexed data store, AVP Record understands media.
If a request is made for content starting mid GoP, then it seamlessly transcodes only the partial GoPs it needs to, delivering the content blindingly fast and only using processing power when absolutely necessary.
AVP Record stores each component of a media package separately (each video, audio, metadata stream in their own fMP4 files). This enables on-demand creation of any content combinations (“give me the main camera video, the German audio and the European region SCTE35 markers…”).
It even enables on-demand creation of trick-play streams (fast forward, rewind, IDR only etc).
Its OpenAPI means customers can build their own front end onto AVP Record for automated highlights, editing in the cloud etc.
One of the most common use-cases is editing live streams and EDL execution and AVP Record comes with an out of the box, frame accurate HTML5 cutting UI.
AVP Record can also form the foundation for an efficient caching / large scale content delivery ('CDN') capability, with the ability to share content between peers, build hierarchical distribution pyramids and use multicast and peer to peer technologies to minimise the amount of data that has to flow over the network.