When broadcasters shift to media over IP, they merge two worlds: audiovisual/media and IT. In doing so, they increase the complexity of their infrastructure tremendously — to the point that traditional management methods involving heavy human interaction are no longer viable.
Prior to the IP migration, a broadcaster’s workflow effectively stayed at the application layer because the network layer was straightforward. After all, SDI is very deterministic. SDI cables are run here and there, and maybe there is a switcher or other device in the middle. It’s just not a dynamic environment.
Following their migration, broadcasters must deal with an infrastructure layer, the IP network, that makes decisions based on a programmed series of conditions. And those conditions change. If not now, then at some point in the future, the behaviour of the infrastructure layer might change based on time of day and day of the week, as well as other factors. These might also include issues, such as congestion or denial-of-service attacks, that can affect network performance or security.
Despite the added complexity of IP-based operations following the merging of AV and IT, the broadcaster must connect these two sets of devices and components so they can work together to fulfil a mission, such as a live broadcast. To make this happen, broadcasters are necessarily turning to automation and orchestration.
Automation
Geared primarily to the infrastructure layer, or IT layer, automation helps broadcasters to address and overcome the complexity of dealing with smart, addressable devices that can advertise their capabilities. For the infrastructure to support operations effectively, thousands of elements must be provisioned, and it’s not humanly possible to keep track of the immense volume of IP addresses and addressable names involved.
When broadcasters connect these addressable elements, the first thing that they do is find that first layer that gives them permission to participate in the network. After the capabilities of an element are advertised, it is “connected” to the application layer.
Automation is the only practical way to tackle that challenge — and to avoid the issues that would inevitably arise from human intervention and error. The task of connecting the infrastructure layer to the application layer is enabled by intrinsic intelligence, and the results of these connections are displayed for operators on a control console dashboard.
Orchestration
Working at the application or media layer, orchestration leverages automation to ensure that the IT side of things properly supports signal flows and needed services.
Thanks to orchestration, the people working on a project can control and define the many elements of the production — receivers, microphones, cameras, and other sources, all providing different angles and shots — without having to deal with what’s under the hood, the IT infrastructure.
This model effectively decouples users (with the exception of IT staff) from knowledge of how the network is set up. It’s so magical, almost nonexistent from their perspective, that they don’t even need to think about it. Through this decoupling, production and broadcast professionals gain greater freedom in exercising creativity and in developing content that will engage audiences. In short, they can create a better and more valuable product.
Monitoring, Diagnostics, and Troubleshooting
The standards used at the application layer are critical to making all of this work. They include not only SMPTE ST 2110 and other standards, but also the IT standards dictated by the IETF and IEEE. While many of them have been around for a while and have become familiar, vendors making IP-based systems for broadcast can make different choices that affect interoperability at the networking level. Thus, for the broadcast engineer, it may seem that there is an issue at the application layer when in fact the problem stems from a fault in the fundamental networking infrastructure.
This is why monitoring, diagnostics, and troubleshooting capabilities are critical for broadcasters working in IP. With intelligence comes complexity — more knobs to be adjusted and more signals to be monitored and verified. Broadcasters need smarter, better tools because they’re no longer just looking for the absence or presence of a flow or signal. They must be able to consider many other nuances pertaining to utilisation of a resource at the IP level.
Such nuances might include utilisation of a link, how much bandwidth is available in that link, and the bandwidth in that link based on what is being put through it. Consider an audio VLAN, for example. Audio really should not consume that much bandwidth compared to video. So, if ports handling audio suddenly are flooded and maxed even though a gigabit or 10 gigabits of bandwidth are available, something has clearly gone wrong.
Ideally, the monitoring dashboard will reflect not only absence or presence of flows and signals, but also information on resource utilisation and other factors that indicate network health. Such tools make it possible to see trends or patterns of degradation that often precede larger failures. Predictive alerts thus can alert operators to take appropriate action. While IP networks never will be deterministic, sophisticated monitoring tools can help broadcasters make their new infrastructure more visible and predictable.
Conclusion
As broadcasters unite AV and IT in support of modern production workflows, they face new and nontrivial challenges. Together, automation and orchestration go a long way in reducing the complexity of IP operations. When backed by sophisticated monitoring and diagnostic systems, they simplify network configuration and maintenance, eliminate the error introduced by human intervention, and give users the freedom to focus more on media workflow than on the infrastructure and connections enabling it.
