By Benjamin Larson, Pre-Sales Engineer at Kollective
Part 1 in a 3-part series about what impacts your live video experience.
We all know when we’re having a bad experience with video. Maybe it’s DirectTV on a windy day where we see pink blocks or smearing in the picture, or a video conference where the person on the other end freezes or even drops. Macro-blocking, lip sync, stuttering, smearing, freezing – the list of ways video can fail goes on and on, but it all means the same thing: your communication is interrupted, and whether it’s changing the channel or picking up the phone, you must do something different.
The scary thing about low-quality video in the context of corporate communications is that not only do your messages fail to reach your audience, your audience starts to tune you out due to the sheer level of friction they encounter while trying to receive your messages. For this reason, it’s helpful to understand what’s going on and what can be done to fix it.
What is Packet Loss?
Modern video communication is digital, composed of millions of packets of information transmitted through some medium. Whether that medium is radio waves or a copper wire, most of the issues we encounter are caused by the same problem: packet loss. When the packets containing the information needed to play back the video get lost, our devices and software don’t have the information required to render the pictures and sound that compose the video. But how are packets lost? Packets are lost when data travelling across a network fail to reach their destination, and this type of data loss is normally caused by network congestion.
Networking hardware such as gateways, routers and switches, used to construct corporate networks, is designed to drop packets when it gets overloaded. Normally, this is not a problem since the TCP/IP protocol is designed to handle that by detecting the missing packets and requesting the server to resend them.
The problem, however, becomes more apparent in the case of time-sensitive, live video communication. Often, by the time the server receives and responds to a request for missing packets, that frame of video or audio sample has already been played. This is especially the case when a router is saturated. The effects of this are all caused by the video player trying to render a picture or play audio with incomplete data.
The problems associated with packet loss and network congestion are compounded when companies try to use video in the workplace, especially for large, one-to-many events such as all-hands meetings. The amount of data required to stream a single live event to all worker’s desktops simultaneously can quickly saturate the amount of bandwidth available on the gateway/firewall for that office. Worse, in addition to bad video, business critical applications are put at risk since their access to bandwidth over the internet becomes limited as well.
What You Can Do About It
You can add more bandwidth, but that’s not the best option because of the cost and complexity of upgrading business critical infrastructure. Additionally, it’s extremely challenging trying to keep pace with ever-increasing demands that video applications place on a business’ network infrastructure. Hardware-based caching appliances have the same cost and complexity, plus the added cost of ongoing maintenance.
Kollective offers a novel software-based approach that is easy to deploy and inexpensive compared with these traditional hardware-based solutions. Kollective’s software-defined enterprise content delivery network (SD ECDN) takes all the bandwidth demands that video applications place on the corporate firewall and pushes them down on to the office LAN where bandwidth availability is not a problem. Instead of all users pulling the same stream through the firewall, only one copy of the stream is pulled per location.
From there, the packets are distributed through an intelligent, distributed network of nodes formed by the software service installed on computers requesting content. Frequently, 95% or more of the bandwidth needed for a given event is reduced. Not only does this give all viewers a much more reliable and higher quality experience, it gives producers a lot more flexibility in how they deliver their video content.