The video best practices in AT&T’s Video Optimizer

The newly rebranded Video Optimizer, formerly Application Resource Optimizer (ARO), still has all the great features found in ARO, but it also has some cool new features, too. The most prominent additions focus on video streaming. The new video features are a big part of the rebranding.

I thought I would share insight into the new video features, some of the thought behind our work developing these new features, and what’s coming next.

Why the Emphasis on Video Streaming?

Video traffic has been growing on wired and wireless networks.

Actually, that’s not accurate. Video traffic has been exploding.

Video on mobile devices is now how people get their news, watch movies, catch up on favorite television shows, follow sports, see concerts, watch commercials, and share the adventures of their cats. Video is the fastest growing content on mobile networks, and AT&T is working hard to deliver the best video experience possible.

In the AT&T Developer Program, we have been working for some time to support mobile developers, content providers, and video producers with new tools that will improve video streaming.

Video Optimizer Video Features

In January of 2017, we released eight new video streaming features that track the following:

Stalls: Stalling occurs when either a user’s device or their network cannot keep up with a video file when streaming. This results in a total pause of video playback

Startup Delays: Streaming video requires a startup delay for smooth delivery. In order to manage buffer occupancy, it is important to understand the startup delay and determine a way to cover this delay for the user with messaging.

Buffer Occupancy: Buffer occupancy is the amount of video stored in RAM to help prevent interruption due to transmission delays, known as “buffering.”

Network Comparison with Current Video Stream: Deliver video at a rate within the network capability, while factoring in the HTTP/TCP protocol behavior.

TCP Connections and the Video Streams: It is a good practice to maintain a reasonable number of TCP connections to deliver a single video.

Video Segment Size: Streaming chunks vary in size. Understanding the size of chunks in a video stream can help to determine the most efficient chunk size.

Video Segment Pacing: Understanding the pacing of chunks in a video stream can help to determine the most efficient delivery.

Video Segment Redundancy: HTTP streaming generates multiple versions of the same content in different quality, which allows the client to display the most appropriate version. Understanding the number and quality of these versions can help a developer avoid overkill.

That’s Just the Beginning for AT&T Video Optimizer

We have a lot more planned to support video streaming, with new features and new video Best Practices guidelines we plan to add to the AT&T Video Optimizer.

You can learn more about other mobile development topics in our Mobile Development Best Practices recommendations.