AV over IP has definitely arrived as a prime distribution platform for AV content. The proof: According to Futuresource Consulting, AV-over-IP encoder/decoder sales were up more than 130 percent in 2017, compared to the year before. A big jump in sales corresponds to a big jump in AV-over-IP installations.
“This is a game changer for the industry and its impact cannot be overstated,” said Anthony Brennan, research analyst at Futuresource Consulting.
Meanwhile, the amount of AV content being carried over IP networks keeps growing. At the rate things are going, “It would take an individual more than 5 million years to watch the amount of video that will cross global IP networks each month in 2021,” according to the Cisco Visual Networking Index: Forecast and Methodology, 2016–2021 report (published 2017). “Every second, a million minutes of video content will cross the network by 2021.”
For AV managers today, moving AV content onto their internal IP networks appears to be an irresistible trend. Even without the push by AV equipment vendors to support IP-based distribution systems, adopting AV over IP allows managers to resolve distance-based signal degradation, switching complexity, and other limits associated with traditional AV distribution systems.
This said, moving AV signals onto an enterprise’s IT infrastructure is not simply a matter of exchanging one form of carriage (traditional, AV-only networks) for another (IP networks). The reason is that IP networks were never designed to carry uninterrupted linear AV content. Instead, the IP standard was optimized to move data files without too much regard for network slowdowns and latency issues. This is because the end user didn’t need access to these files until the download was complete.
The fact that AV over IP requires the transmitted signal to be accessible and usable throughout the entire transmission—especially when the AV content is being broadcast in real time for live viewing—imposes a requirement that traditional data applications didn’t have. Given this fact, there are a number of issues that AV managers wishing to move to IT-based content delivery need to address to make this transition successfully.
KEEP AV ON ITS OWN IP NETWORK
The experts interviewed for this article agree: Whether virtual or physical, AV content should be moved over its own IP network, rather than mixed in with non-AV content on the main enterprise network. “In many of the integrations involving our ZyPer4K products, AV operates on a dedicated network so that bandwidth and security is not compromised,” said Art Weeks, ZeeVee’s product manager for IP products.
There are many reasons to keep AV content separate from non-AV content. First, keeping the two apart eliminates the issue of prioritizing one form of traffic over the other—especially if the shared network is also being used to carry Voice-over-IP (VoIP) telephone calls that cannot tolerate any degree of latency or signal dropout.
Second, keeping AV segregated from non-AV traffic is a good security move. “Many of the AV-over-IP connection boxes are just simple Linux devices that can be relatively easy to hack,” said Phil Hippensteel, assistant professor of information science at Penn State University. “Having these boxes connected directly into your main IP network could make highly sensitive corporate data vulnerable.”
Security issues related to AV content depend on the specific application. “For instance, security may not be an issue in a sports bar, where it would likely be a concern in a corporate environment,” Weeks said. Nevertheless, “the security of the AV network itself is not the issue. The real issue is that you do not want AV aspects of the network used by outsiders to gain access to the non-AV portion of the network.”
A third reason to keep AV separate from non-AV traffic is to prevent AV content from devouring the available bandwidth on a shared network. This can be a problem with networks that use the Internet Group Management Protocol (IGMP) on IPv4 networks to multicast to numerous users.
“What happens with IGMP is the switch isn’t intelligent enough to know what its limits are when streaming AV content,” said Chris Fitzsimmons, a product manager with Biamp Systems. “Should the stream exceed the available bandwidth, we’ll start to see packets of data tossed over the side.”
This issue can be averted by using Audio Video Bridging over Ethernet (AVB)-based streaming instead, which reserves 25 percent of available bandwidth for non-streaming traffic, Fitzsimmons said. But again, a simpler solution is to have AV content on a separate network (virtual or physical) instead.
CHOOSE TCP OR UDP, DEPENDING ON YOUR NEEDS
One important detail to remember with IP networks is that the standards that dictate how data should be packaged and prioritized—the transport protocols—are definitely not one-size-fits-all solutions. Instead, each transport protocol has pros and cons, in line with how it moves data across the network and what priorities it addresses.
For AV content transmission, two common options are Transmission Control Protocol (TCP) and User Datagram Protocol (UDP).
If you select TCP, “then your AV stream will grab all the bandwidth it can grab; even it if maxes out the network,” Hippensteel said. “If you select UDP, then the amount of bandwidth dedicated to your AV stream will remain constant. The good news is that this will keep the network from being maxed out. The bad news is that if the UDP’s bandwidth has not been configured to provide sufficient capacity for the AV stream, the signal may stutter and/or degrade in resolution.”
KEEP AV IN A SINGLE DOMAIN
When it comes to IP traffic, the domain name of the content is akin to the address of a house. This is usually a word—i.e. Google.com—that is associated with a numerical IP address or addresses that internet routers use to connect a computer to that network location.
Since a domain can support more than one IP address, it makes sense to keep things simple by having one domain for users accessing AV content. But that’s not the only reason to keep things simple: “As a general rule, you want to keep all AV-related products on a single domain as control is difficult to maintain over multiple domains,” Weeks said.
GET THE RIGHT CODEC
A codec (coder/decoder) is the key piece of software that encodes and then decodes AV content for transport and viewing over an IP network.
Like the transport protocols, there is no universal solution when it comes to choosing the right codec. “It all depends on the use case,” Fitzsimmons said. “Are you doing digital signage that draws on recorded content, or a keynote speech that needs real-time delivery with minimal latency? Your use case will determine which codec is right for you; ask your IT support person if you’re not sure.”
THE BOTTOM LINE
Moving AV over IP is not just a case of switching copper AV cables for Ethernet cables. Make sure to do your homework—and to work closely with your IT department—before making any purchasing decisions.
(Note: For a quick primer on AV over IP, check Matrox’s “Fundamentals of AV over IP” webpage at http://www.matrox.com/graphics/en/press/guides/av-over-ip-fundamentals/)
Considerations for Moving AV over IP
- Keeping AV traffic separate from non-AV traffic (either on a virtual or physical network) minimizes traffic jams and reduces security risks.
- Know what content you want to move, and how it will be used (e.g. live broadcasts versus digital signage) before choosing codecs and transport protocols.
- Your transport protocol of choice (TCP or UDP) depends on what you what your AV content to do, and how it will be consumed.
- Moving to IP is complex; get help!
- Buy-in from your IT department from the outset maximizes your chances for success, and peaceful coexistence.
