By Kevin Crayton – VP of Product Management
An enterprise video strategy offers organizations a unique opportunity to leverage more immersive communications tools to engage employees, train the workforce, spread corporate messages and hold virtual meetings. But video also creates technological challenges, particularly when it comes to delivering content through the network.
Most corporate networks are designed to deliver a large quantity of small data packets through the WAN, carrying them into the LAN, which generally boasts much more bandwidth. When video is sent through the WAN, a single data packet can use all of the available bandwidth, creating performance problems as other packets related to the content and non-video data, such as application-related packets, are dropped and cannot get through the network.
Because video data packets are so large, just adding bandwidth is not a viable solution. Instead, organizations need to analyze the types of video they plan to use and deploy a video-specific solution that meets their specific operational demands. While there are a variety of video delivery options available, most are built using either multicasting or peer-assisted methods. Choosing between the two based on operational needs is vital to getting the most out of your plan.
Often featured within enterprise content delivery networks, multicasting enables video to be delivered using as little bandwidth as possible by sending content through the WAN in a single data packet. Typically, multiple users trying to sign on to a webcast or view content in a similar way will be doing so at the same time, leading to multiple large data packets trying to go through the WAN simultaneously. Multicasting overcomes this by programming the system to recognize when multiple users are trying to access a single video, compressing the data into a single packet, then sending it through the WAN using much less bandwidth. Once the data packet has arrived within the LAN, it is then split up and delivered to the various users viewing the content.
From an operational perspective, multicasting is often ideal if you plan to have a number of viewers watching the same content simultaneously because it ensures that live stream events, webcasts or other types of video get through the network with minimal bandwidth disruption. However, it can also be incredibly difficult to configure properly because an effective multicasting solution depends on major adjustments to the network setup.
Peer-assisted video delivery
Like multicasting, a peer-assisted solution focuses on alleviating the amount of bandwidth traveling through the WAN. However, it accomplishes this by only sending the data pertaining to a video once, then having it rest as cache within the LAN until everybody who is supposed to watch it has done so. While this sounds simple, it can actually be incredibly complex to configure, as each piece of content has to have permissions, routing information and details as to who is supposed to watch it attached to it before being sent through the network.
However, some leading vendors have developed methods to simplify the configuration of peer-assisted solutions. As a result, peer-assisted delivery methods are often well suited for organizations that plan to create content on demand and then have viewers watch it within a certain time frame, not necessarily simultaneously.
Choosing the right solution for your needs
At their core, multicasting and peer-assisted delivery methods do the same thing – reduce the impact video content has on the WAN. However, each delivery method works best with specific types of content. As a result, organizations hoping to get the best performance possible for their needs have to consider how they plan to use video and deploy a delivery solution that will match those operational requirements. Without this planning, performance may improve, but not necessarily to the degree needed to support ongoing requirements.