Redundant Versions of Same Video Content

Introduction

Adaptive Bitrate streaming (ABR or ABS) generates multiple versions of the same content in different levels of quality, and allows the client to request the most appropriate versions.

Understanding the number and quality of these versions can help a developer avoid the overkill of creating and storing unnecessary versions of the same content.

Background

ABR works by dynamically monitoring the CPU and memory capacity of a device, and then making corresponding adjustments to requests for the next video file.

The server stores multiple versions for every single video clip, each with a different level of quality, and sends the files that the player determines will play at the highest level.

The process involves encoding the source video at varying bit rates on the server, and then segmenting each of the different bit rate streams into small parts, or segments. An encoder processes a single source video at multiple bit rates, and the player switches between streaming the different encodings depending on available resources. The result is very little buffering, a faster start time, and a good experience for both high-end and low-end connections.

The segment length typically varies between 2 and 10 seconds. The user’s multimedia player can take advantage of ABR if it has the capability to switch among the different bitrate segments—locating the segments that correspond best to the bandwidth on the user’s computer.

The Issue

The question is how many separate versions should be created for each video file?

More versions will give the client more flexibility to choose the best quality based on the available bitrate and display capabilities.

However, each different version that gets sent and never used adds additional overhead to the network, slowing down transmissions, increasing latency, and increasing computing and storage costs, without being consumed by the player or seen by the viewer.

Best Practice Recommendation

The Video Optimizer can help you identify and compare redundant content sent as additional versions of the same content. This will help you determine the optimal number of different quality versions assigned to each video clip.

The Video Optimizer compares the number of redundant segments downloaded to the number that were actually used. Tracking the redundant versions can help you manage them better.

Video and other types of rich media are on the rise in mobile apps and in the future we plan on offering more guidance on streaming and other video technology in our Mobile Development Best Practices recommendations.