One of the biggest challenges for Pay TV operators today is how they retain and grow their subscriber base. If they get into the habit of introducing compelling new services, like 4K/UHD, some time before market rivals, that is going to give them a competitive advantage and help in this endeavour. And this ability to ‘get out of the gate first’ is enhanced by the move to software-defined video and then a unified headend.
Dan Gehred, Solutions Marketing Manager, Compression at AWS Elemental, which pioneered high-performance software-based video processing, says there is definitely a first-mover advantage for companies that unify the video processing platforms and some of the associated infrastructure needed for what are currently separate broadcast (or primary screen) and multiscreen (including TV Everywhere or other OTT services) video workflows. “The sooner you move to a unified headend, the sooner you can take advantage of its flexibility and efficiencies,” he says.
So what precisely are the benefits from a unified headend? These were discussed in detail during a Videonet webcast this week, together with the technology that underpins a unified headend, the main technology challenges when migrating, the likely impact on your organisation, what it means for redundancy and how you convince a CEO that this should be a priority. You can hear the full 55-minute discussion here, free.
If you are already using software-defined video, where video processing software is no longer tied to specially optimised, single function hardware but instead can be run on generic compute platforms (including appliances), you are already part-way to the unified headend. Gehred noted a key benefit of software video processing: the ability to implement and update features much more rapidly, and without big new capital investments. This innovation could include use of new codecs or enhancements like HDR (high dynamic range) and Wide Colour Gamut.
He also noted how the high performance of software-based video processing has opened the way for the unified headend. “Software-defined video can do anything that specially built, optimised hardware did in the past. You don’t have to rely on an ASIC-based software-on-chip solution today,” he declared.
This means that live broadcasting to the big screen television, including for 4kp60 channels, can be software-based. “This [software-defined video] can handle not only the OTT and TV Everywhere workflow, but the primary screen encoding, and that helps create a platform that can be used for both cases.”
At the same time, there has been some convergence around the choice of codecs, and this is also pointing towards headend unification. “As codecs evolve and new ones arrive, the gap has closed between the codecs that are used for multiscreen and what is used for the primary screen,” Gehred said.
With MPEG-2 very much a legacy codec on the broadcast side, H.264 and HEVC have their future across both workflows. “As the video compression you need gets more similar, this is helping to push operators and broadcasters in the direction of combining their operations,” Gehred told the live webcast audience.
When you make the move beyond software-defined video and into a unified headend there are some immediate benefits, which Gehred listed:
- Less equipment, less cable, reduced rack space, lower energy and power requirements – all leading to lower costs
- The ability to achieve full system redundancy at lower cost once you have combined all video processing into a single solution (less spares needed than before you unified)
- Centralised operational monitoring, which brings its own efficiencies (it is easier to provision new channels or services, including temporary ones)
- Networking efficiencies once you reduce video traffic routing requirements (especially if you have less feeds coming in to disparate locations)
- More room in your headend to expand other services.
Jamie Duemo, Senior Product Manager at AWS Elemental added: “Because you collapse two video paths into a ‘one input and many output’ methodology, the first benefit is reduced monitoring and alarm points upstream and you will spend less time chasing false-positive alarms and less time trouble-shooting when there is a real alarm.
“That means you can use your operations and engineering teams more efficiently. Overall, the best part of having a less complicated architecture is that it is easier to understand, you need less documentation, less effort keeping up with documentation and less time spent on training. Engineering, design and integration cycles can be redirected towards more positive initiatives like R&D, which means more innovation in the end-product you take to consumers.”
She emphasised the powerful impact – which you notice from the outset when you unify headend operations – of increased efficiency, agility and cost savings. She also made it clear that you do not have to commit to a ‘big bang’ unification initiative but can conduct tests using a software-based, shadow (non-production) workflow and use these to learn the technology, train staff, and learn more about how to redefine the workflow and where all the efficiencies are.
“You monitor the secondary [shadow] path like you would the primary [real] path and do that over a period of time to check you are getting the quality and reliability you need. You can discover what upstream business systems might need to be slightly adapted, like some module plug-ins that may be needed. You can decide the key metrics that the business wants to capture and define what success is.”
The Videonet webcast provides more detail about the technical challenges when migrating to a unified headend, including the biggest one – interoperability. It addresses some of the standardisation initiatives that will help make a unified headend easier to implement. And there is a discussion about how BT, the UK IPTV provider, has implemented a unified headend that combines multiscreen and primary screen delivery. You can hear the full webcast, on-demand, here.