When it comes to streaming video, bandwidth optimization is critical. Corporate internet connections and WAN links are usually limited in bandwidth, so streaming video across them impacts network performance.
The more you use video, the greater toll it takes on your network. It can slow busines-critical processes and impact quality of experience—and viewer expectations are high. According to Wainhouse Research, 72% of those watching online business video daily say that “broadcast quality” receives “significant emphasis” when grading the effectiveness of streaming video.1
One of the best ways to reduce the amount of video traveling your corporate network is to move the video source closer to your viewers. The less distance the video has to travel, the less bandwidth it consumes.
Ramp OmniCache is a caching solution designed specifically for video. It is vendor neutral, secure and doesn’t require a client or plugin on viewing devices. OmniCache also includes advanced bandwidth optimization to ensure network stability.
Bandwidth Optimization: Delivery Profiles
Today’s most common video streaming protocols, HTTP Live Streaming (HLS) and Dynamic Adaptive Streaming over HTTP (DASH), are both adaptive streaming technologies. They are designed to serve the best quality of video at any time based on the bandwidth available and capability of the viewing device.
To do this, they format their videos as a package of individual video, audio and closed captioning files called tracks. The video tracks differ in quality, resolution and bitrate to cater to variations in internet bandwidth. Audio tracks are typically used to support multiple language translations. Regardless of how many tracks are used, OmniCache effortlessly stores and redistributes these tracks. Any time a repeat request is made, you get bandwidth optimization.
However, the circumstances that make adaptive streaming desirable on the public internet are less applicable on a corporate LAN. On the LAN, you usually have enough bandwidth to deliver the highest quality video to all viewing devices. In fact, by not catering to adaptive streaming, the corporate network can recognize even more bandwidth optimization. This is because only one version of the video track is requested from the video source, stored and redistributed (as opposed to multiple versions).
OmniCache includes a configuration option that allows you to select a single video track on behalf of the player. When you choose this delivery profile, OmniCache rewrites the playlist/manifest to exclude all other track options, creating bandwidth optimization.
You can also change the delivery profiles at locations with low capacity internet or WAN connections. Instead of selecting the highest quality video, you choose an optimum level for each location. With this type of control, you can set the delivery profiles to ensure bandwidth optimization across your network.
Bandwidth Optimization: Collapsed Forwarding
When OmniCache gets a request for a segment that’s not stored in the local cache, it requests the it from the source. The source then returns the video segment to the local cache, and OmniCache distributes it to the viewer. This process takes some computing time, but not enough that you or I would even notice.
During a live stream, it’s common for many viewers to request the same video segment at the same time. As a result, the cache forwards all requests until the segment is available in storage. The problem is these repeat requests erode the eCDN’s bandwidth optimization.
OmniCache has a built-in capability called collapsed forwarding that prevents this issue. OmniCache knows if a viewer has already requested a video segment and holds all subsequent requests until the segment is available.
Since it’s requesting video segments before they are needed, viewers don’t experience any delays and less WAN bandwidth is consumed.
Bandwidth Optimization: Live Prefetch
Unlike video on demand, where the entire video is complete before segments are available, live streaming creates video segments as the video is recorded. Because most people are watching at about the same point in the video, their players are requesting the same segment at the same time. This leads to a lot of requests for the same segment at once.
While collapsed forwarding definitely improves this situation, it would be even better if you didn’t have a constant loop of requests for video segments from the source.
OmniCache’s live prefetch feature predicts the next segment and attempts to retrieve it in anticipation of the next request. It understand how segments are numbered and when they are likely to be available. Live prefetch maximizes the likelihood the segments will already be available in the cache.
Live prefetch combined with collapsed forwarding gives OmniCache the best opportunity to serve video segments from the cache. This maximizes bandwidth optimization without compromising user experience.
Want to learn more about OmniCache?
OmniCache is a robust video cache purpose-built to deliver enterprise video with the lowest possible impact on your network infrastructure and security. While other eCDNs provide the most value during live events, OmniCache optimizes both live and on-demand video. To learn more, check out the following resources:
- Read our OmniCache Technical Whitepaper
- Get the OmniCache Solution Brief
1 Survey Insight: Gauging the Effectiveness of Video as an Enterprise Communications Tool
