Ensuring Live Streaming Achieves Broadcast-Grade: Part 2 - New Opportunities

As specialist broadcast service providers and network operators look at the fast growing requirement for at-scale broadcast-grade streaming video, it is a big opportunity for them to address.

The last 10 years have been mostly about VOD and binge-watching, live that is “good enough” mostly on smaller-screen mobile devices, and a side-business of broadcaster D2C or “TV Everywhere” that is soon set to become the main business. The next 10 years are a big opportunity for the broadcast industry service providers to become big streaming players.

Already these service providers provide broadcast-grade contribution and transmission services to their broadcaster customers. Now they want to add streaming services to their portfolio. And why not? CDN services and solutions exist that they can utilise in new and economically scalable ways compared to the traditional public CDN resell model. ISPs could benefit from specialist partners that aggregate and manage the big streamers who represent such a high percentage of total internet bandwidth consumption. And consumers are moving rapidly to streaming as the preferred platform. So, broadcasters need to ensure their services are of the best quality. Which always means “broadcast-grade” because what else defines “best quality”?

Broadcast service providers that specialise in live video delivery know that they have the operational capability to deliver broadcast-grade services. They need to wrap their arms around the CDN platforms, the ISP relationships, and the player and client analytics, to take as much control as they possibly can of the end-to-end video delivery chain. With this control they can streamline and optimise delivery efficiency, and proactively manage quality. Exactly what the broadcasters require in order to satisfy their quality-expectant audiences and monetize their content.

Looking at the full value chain, already the broadcaster service providers know how to deliver low-latency and high quality. Their first rule is that the end-to-end chain must be set up for success. Upconverting a stream part way through the chain is not a good idea. Progressive content is a must, not interlaced. While you can convert interlaced to progressive, it is best to capture content in progressive. Second, it is best to have dedicated fibre connectivity from point of capture to point of origination. Internet protocols like SRT, Zixi and RIST are helpful for lower budget productions, but when live must mean live these protocols add too many seconds to end to end latency. What we first want is point-to-point fibre transmission with intra-frame encoding like J2K/J2XS for lowest delay and highest quality.

Figure 1: The E2E video delivery chain, encompassing traditional networks and CDNs.

Figure 1: The E2E video delivery chain, encompassing traditional networks and CDNs.

At this point broadcast-grade encoding is necessary to produce consistently high quality and low latency outputs. Then, for high bitrate streaming to bigger screens, low-latency HLS/DASH/CMAF are best – they achieve highest quality at lowest latency. CDN-based delivery of these streams to the receiving Players is required to handle the chunked video fragments to the consumer. This is the dominant model for streaming today and into the future because it sustains the highest quality, which we need on our larger screens and for enjoyable viewing experiences.

When very low latency is required (<1 second), but lower video quality is acceptable, WebRTC (Web Real-Time Communication) is a good choice. This protocol was developed for video conferencing on browsers and is not compatible with CDN-based delivery. WebRTC can handle all levels of resolution, and it is very fast. But it is not based on video fragments, meaning that a very low level of signal disruption due to network problems (like congestion) can cause a problem with the quality. When consumers request a stream they are directed to the closest WebRTC servers and then receive CBR streams to their devices. If the quality of the connection degrades, the servers will switch to lower bitrate CBR streams. With these two primary content delivery choices, because the internet is not a perfectly consistent system and our viewing enjoyment is generally more important than a few seconds of latency, OTT video uses the chunk-based CDN/Player approach.

Once encoded and packaged the streams are pushed out into the distribution networks. The first rule is that for best performance you cannot use an unmanaged IP network like the internet. You must have a managed network. And managed doesn’t mean hundreds or thousands of D2C broadcasters running in a multi-tenant environment where capacity is difficult or impossible to manage. So what does managed mean? Managed means that individual streams are clearly visible to the streaming operator and are proactively managed to broadcast-grade SLAs during live events. Managed means that the fibre connections and hops in the chain from encoder to ISP are under the tight control of an E2E network operator. Managed means that Edge nodes are under the control of the broadcaster service provider so that traffic can be handled, and capacity can be deployed and upgraded rapidly to support D2C service growth.

The CDN component is a major component in this chain. Rather than hand-off to multi-tenant, multi-purpose public CDNs on a pay-as-you-go basis, the leading broadcaster service providers are looking to create purpose-designed video delivery networks. These video delivery networks will be dedicated to live video delivery first, with VOD (or rather catch-up TV which is more relevant for live video) as a natural secondary use case. As we know, live sports and live shows are the big drivers of the largest audiences, and catch-up follows for a short period afterwards. VOD rumbles on all the time but doesn’t drive the biggest peaks of viewing. Plus, from a network management perspective and service management perspective, VOD is easy and Live is hard, simply because of peak capacity consumption and what this means for video delivery quality in a congested and dynamic network environment. This is the basic reason why broadcast service providers offering purpose-built live video streaming services can best solve for 4K+ prime-time live video delivery (See Part 1). They know best what is required to get the video through the pipe.

The ISP relationship is quickly becoming a much more strategic relationship for content providers and their service providers. ISPs are fundamental to the delivery of high-quality video. Edge caches need to be as close as possible to the consumer and the network topology must be designed carefully for live event surges that appear in different geographies (the classic situation is the local sports derby when audiences are more densely populated in particular locations). These situations require ISP collaboration. Ideally the Edge caches are deployed deep inside the ISP networks, not at peering points. Network routing tables should be configured for the expected traffic and should be dynamically adjustable during live events according to the audience behaviour. This becomes even more important as video resolutions increase and the video consumes more bandwidth. The vision is for the broadcaster service provider to become a live video hyper-scaler, representing the collective interests of the small number of broadcasters who are big bandwidth consumers for the ISPs. This approach could also bring benefits to the ISPs who can consolidate operations, reduce core network capacity usage, and optimise their network design to handle 4K+ OTT video.

Broadcast-grade also means having SLAs that provide high-levels of video delivery performance and service availability. This results in broadcaster service providers having 24x7 operational teams with experts on call and swift escalation processes. It also means having resiliency and redundancy built into the network design, so that peak traffic can be managed even if there are significant network outages. Fiber cuts and maintenance outages are notoriously awkward and unavoidable, so systems must cater for these types of unexpected outages, rerouting traffic with no notice. Applying the same design and operational approaches to streaming is essential in order to reach broadcast-grade.

Robert Szabo-Rowe, Senior Vice President of Engineering & Product Management at The Switch states, “For years we have worked with some of the world’s largest live events, ensuring high quality delivery from camera to Headend or Origin Server. Today our customers are ramping up their D2C business and demanding parity of experience between delivery platforms. We believe we can assure quality of delivery to the consumer, leveraging all our experience in video network management for the most demanding audiences”.

In the end, the demand for high bandwidth streaming is coming. It won’t be stopped. The wave already started and is building up, and the tsunami will crash heavily onto the shores of the ISP networks unless there is a change in how we manage the delivery of the video. The Super Bowl is expected to drive about 100 Tbps of bandwidth in the US in 2023. Offering 4K to subscribers on a premier streaming platform in one European country is expected to triple the size of the CDN required to reach about 30 Tbps. These are colossal numbers compared to what we are used to, and they are growing much faster than internet capacity is growing. It needs managing.

Broadcaster service providers who know how to manage live events and live video can respond to this challenge, adopting new technology models that offer a win-win-win for themselves, the broadcasters, and the ISPs. Ultimately, the consumer benefits as these three major industry players can together ensure that the highest possible video quality appears flawlessly on our screens.

You might also like...

Vendor Spotlight: Imagine Communications

Imagine supports media companies’ bottom line with “converged” technology platform.

Orchestrating Resources For Large-Scale Events: Part 2 - Connecting Remote Locations

A discussion of how to create reliable, secure, high-bandwidth connectivity between multiple remote locations, your remote production hub, and distributed production teams.

The OTT Lexicon For 2023: Part 1 - Is Our Terminology Getting A Bit OTT?

The world of streaming is defined by acronyms like SVOD, AVOD, FAST, OTT and more. But this leaves gaps and confusion in what is included in our OTT services. For example, what does a service like BBC iPlayer include? What…

Orchestrating Resources For Large-Scale Events: Part 1 - Planning Is Everything

An examination of how to plan & schedule resources to create resilient temporary multi-site broadcast production systems.

Sustainability Of Streaming: How Does OTT Compare With OTA? - Part 3

Parts 1 and 2 in this 3-part series analyzed the latest information about OTT and DTT energy consumption in the UK in the year 2021, concluding that there are important energy efficiency improvements to make in OTT, and some big decisions coming our…