AWS Media Blog

The Evolution of the Television Live Event Ecosystem and How AWS and TAGVS are Shaping the Future – Part 1

This blog was co-authored by Kevin Joyce (TAG Video Systems), Peter Wharton (Happy Robots), Michael Proulx (Retired CTO), and Mark Stephens (Amazon Web Services).

This blog is the first of a two-part series in which we walk through the linear television live event ecosystem and how it is evolving through the use of cloud. In part 1, we walk through what the ecosystem looks like today. We cover the delivery process , the challenges being faced by the industry, and how they have scaled event production on-premises. We also talk about over-the-top (OTT) delivery and the complexity of ensuring that the end-to-end ecosystem is performing the way you expect.

The Live Event Ecosystem Today

Linear television is the delivery of video content to consumers on 24/7 channels typically as broadcast or cable channels. Live events, especially sports, are an increasingly important part of the linear television model as they have the value of immediacy that VOD and binge viewing lack. As a result, live events generate an significant share of the advertising revenue and play an vital role in sustaining subscriptions and ads to channels and services.

Figure 1 illustrates the overall ecosystem of live event production and delivery today

Figure 1 illustrates the overall ecosystem of live event production and delivery today. There are three key processes in the ecosystem: live production, playout integration, and delivery. Next, we will briefly describe each of the three processes.

Live Production Process

In the live production process, the live show is created by switching between various live action cameras intermixed with replays, commentary, and other content. Live event statistics and graphics are also overlaid at this point.

Figure 2 illustrates the evolving models for live production.

Figure 2 illustrates the evolving models for live production. Historically the live production process was performed at the event venue either in a permanent production facility or, more likely, in a mobile production truck/Outside Broadcast (OB) van. Recently broadcasters have deployed a new model of “remote production” in which the cameras and minimal personnel remain at the venue. The bulk of the equipment and production personnel remain in the broadcast facility, where all of the camera signals are delivered. NBC and other international broadcasters pioneered this process a number of years ago for Olympic Games coverage to save on production costs. Many other broadcasters have since applied the model to a broader range of events and remote production has become one of the most important trends for live production.

More recently, production service providers including traditional production truck companies have started to build hubs for remote production. These hubs concentrate core production equipment and personnel at central facilities with multiple live production control rooms that can simultaneously produce numerous live events. In some cases, the production personnel operate from control rooms or galleries at the hub; in other cases, the equipment is in the hub but the personnel operate from a remote gallery. The likely next step in this evolution will be the ability to operate remote production using public cloud-based infrastructure, enabling operators to scale up services to match the number of live events they need to produce.

Playout Integration Process

The second process involves the integration of the finished live event produced during live production by adding commercials, promos, station IDs, and other interstitial programming. SCTE “markers” are inserted for downstream ad insertion and localized channel branding graphics. This process is typically performed by the broadcaster who has acquired the rights to the event at the broadcaster’s facility or at the playout service provider’s channel origination facility.

Delivery Process

The third process we’ll describe in this blog is the delivery an integrated program to viewers and subscribers. Historically this distribution has been mainly through conventional TV service providers including terrestrial (over-the-air) and Pay TV operators such as cable, IPTV, and direct-to-home satellite service providers.

The delivery process has grown to include OTT and digital distribution, during which linear streams are delivered to users over a public Internet connection.

Figure 3 illustrates the difference between cable tv delivery and OTT delivery of content via the internet.

Figure 3 illustrates the difference between cable TV delivery and OTT delivery of content via the internet. OTT delivery will involve an origin, CDN and edge infrastructure for delivery to internet connected devices.

Challenges in the Live Production Ecosystem Today

As live linear channels increasingly depend on live events, the number of events and the number of outlets continues to increase. There are a number of challenges in the existing live event production and delivery ecosystem:

  • Production equipment and personnel underutilization. Remote production and production hubs are helping to mitigate this issue, but as more events are produced, particularly lower revenue, niche audience events, more dynamic and agile models will be needed to deliver the necessary economics. Orchestrating these systems end-to-end from event acquisition through transport, production, playout, delivery, monitoring, viewer analytics and monetization is a daunting challenge.
  • The static and non-optimal use of conventional Pay TV delivery. The infrastructure of conventional Pay TV operators is heavily hardware-dependent. This results in a system that supports a limited and often static number of channels. As live event coverage scales, more dynamic models are needed, including temporary “pop-up” event-based channels. Operators are challenged with making sure that these temporary channels perform as well as their full-time equivalents, including edge-based and server-side ad insertion and regional blackouts.
  • OTT delivery leverages public networks for both contribution and final delivery to users. The infrastructure is complex with many different components often distributed in different locations. Additional care must be taken to monitor the overall system from end to end. Specifically for high value event, the monitoring must include an “Eyes on Images” element to make sure the proper content is being delivered and the Ad insertion markers are present. Lower tier channels require highly automated solutions that deliver similar results.
  • The network and components in an OTT delivery system add a substantial amount of latency to the overall program. For high-value live events, latency is problematic: consider the increase in live betting and the possible phenomenon of subscribers hearing a neighbor on a different platform cheering for the big goal 20 seconds before they see it.

In part two of this blog, we examine how the cloud is changing live event delivery and why OTT has been the first distribution model to leverage the cloud.