As was the case with many of us adjusting to socially distanced teamwork this year, editorial and production professionals in the broadcast industry had to adapt quickly, and in the case of news editorial, they did so in the context of also having to cover a global emergency. But this transition to remote work is an acceleration of a trend that was already bubbling over the last several years as the use of centralized data centers and cloud resources has become more common in broadcasting, whether to protect intellectual property, to optimize efficiency, or simply to be able to respond to the needs of ever-shrinking production timelines and budgets.
Ian Main, Teradici’s resident expert in how media and entertainment companies leverage remote access technology in their production workflows, recently sat down with James Careless of TV Technology for a more in-depth discussion of the use of Teradici technology within broadcasting workflows. This Q&A is an edited excerpt of that conversation. The full discussion is available for viewing here.
We have lots of interesting use cases in TV production. A good example is Afrokaans Film & Television, which is a South African production of the famous Survivor series. You can imagine a remote television editorial filming use case – you have all the complexities of remote sites and the difficulties of shipping big media disks and files around. At one time the artists and video editors had to fly to the site to do the editorial work. Once they were using Cloud Access Software, they were able to upload that content to a central server location that editors could access from anywhere. The entire pipeline from rough cut editing through to later production could all be done remotely. Now of course, we're talking primarily editorial because of the high-performance use case – editorial has a big graphics demand and so on – but there are lots of other use cases within the broadcast studio where I think we'll have lots of value to add.
Teradici works with many major editorial software providers, like AVID, Adobe, EditShare and other major video editorial teams to make sure we have a compliant integrated solution, but our PCoIP® software itself isn't that concerned with the actual video application. We will work with most applications right out of the box, but we partner with these editorial software vendors so that the end customer gets a great integrated experience.
We want to see that the infrastructure supports the use cases, so we look at things like the network bandwidth between the host environment and the endpoint. For a dual 1080p setup, for example, we recommend 10 to 20 megabits per second, and as they go to bigger displays, more may be required. The interactive latency is another interesting requirement. The protocol supports a wide range of network latencies up to 250 milliseconds, but to be able to work very efficiently we'd suggest around 50 milliseconds or less.
Just to provide some insight into what that means: I'm here in Vancouver, Canada and my workstation that I use every day is down in Portland, Oregon, and that's under 20 milliseconds of latency. So typically, one data center could service the West Coast, another could service the East Coast, and a corporation that's national could support the entire workforce from one or two data centers across the country.
The editor has the same normal display system – the normal peripheral devices, Wacom pen displays and tablets, and so on. What's interesting is they could actually be getting a better experience than local, because once you’re set up for virtualized infrastructure, you’re set up for very high-performance storage and high-performance graphics. So, what happens is studios can invest in that infrastructure to provide that amazing experience compared to say, doing editorial work on a laptop. Once you've got very high performance storage connections and a good GPU that can be shared among users, from an artist's point of view you're able to experience what you're used to on a local system, and perhaps even higher performance because you've got a bigger GPU or CPU offering on the back end of all that.
I think this is a big shift. Broadcasters move slowly because they've got big infrastructure investments, but what they're seeing is all sorts of opportunities to leverage this remote work methodology. Firstly, geographic presence – things like remote news, remote productions for TV, and so on. Then another component is taking on contractors – rather than just investing in more on-premises compute infrastructure, broadcasters can start thinking about leveraging public cloud and this allows them to scale very rapidly up and down.
They can also consider moving people out of expensive cities. We have customers today who have allowed their employees the flexibility – and this has been accelerated by COVID, I think – to start working in less traditional places, so customers are setting up facilities in smaller towns. We have a great example in Jellyfish Pictures out of the U.K., which set up a branch in Sheffield to get away from that expensive London real estate.
Then there’s the CapEx side of this. Broadcasters spend a lot of money on infrastructure, but if they can put at least part of that infrastructure into an OpEx mode, they can respond more rapidly to new programs coming online and the need to accelerate the timing of applications.
We're starting to see more confidence in the public cloud offerings to instantiate virtual workstations and all the public cloud vendors now offer these cloud-based workstations, so now you can expand to any region.
Cloud Access Software has a management plane we call Cloud Access Manager, which is a single management console that allows for this hybrid environment where user entitlements and configurations of cloud machines are managed, whether in public cloud platforms or on-premises data centers. Connections are brokered for users that could be anywhere in the world, effectively, under that single simple management regime.
You can provision machines to be shut down when you're not using them, and you don't pay for these cloud resources when they're not at work, so you can augment your on-prem infrastructure in a hybrid approach and budget on a pay-as-you-go basis.
The other thing that's interesting is that broadcasters are looking to leverage sophisticated high-value services like AI and machine learning to support aspects of their broadcasting workflow – things like language translation, marketing, statistics-gathering. All these things benefit from high performance compute. In those cases it makes sense to put your workstations, or some of them at least, where you leverage those services, using public cloud facilities.
Broadcasters in general are moving away from SDI networks, which have prevailed for a very long time, to IP based networking, and the Cloud Access architecture goes hand-in-hand with this IP migration to gain flexibility. All that previous KVM technology involves proprietary hardware devices that can't easily be upgraded, so you end up with a high ongoing capital cost whenever there's a content format change or an audio standard change – the old hardware needs to be thrown out. Once you move to a software environment, you get this network efficiency and flexibility, but you can also support new content needs without all the extra hardware costs. With PCoIP for example, we're working on various emerging formats such as 4K and HDR support, and these all done now through software upgrades rather than expensive infrastructure changes, so these are these are big things that will help the industry move faster in the future.
I think we're all very aware that there's a new normal now, and they need to respond to this new way of working where at least some people will not be in the office. This is a solution for that, and not only does it solve the immediate problem, but it sets you up for that longer-term migration of broadcast infrastructure.
There's also business continuity planning in general. When we talk about setting up for disaster recovery and these types of things, you can now have desktops or workstations on standby in case you need to respond, or in separate locations that you may need to bring up.
Another aspect to this is access to video editorial talent and visual effects talent. These are sought-after people that are spread around the world, so you can gain access to this talent in your own jurisdiction by simply giving them resources that that they can access from wherever they are.
What we’ve seen is some have just plunged into the deep end with their most urgent use cases - install our Cloud Access Software Agent on the on-premises workstation, install the Cloud Access Connector as the secure corporate gateway and connect using a software client. First address those urgent needs and try it out.
Now for the more substantial migration, you can take a more organized approach. I think you target the teams or use cases that benefit most from this remote access, and then set them up for remote working as a trial. We offer trial software. And, of course, look at their bandwidth needs – what is their latency going to look like, does the protocol support it to your existing broadcast studio, or does it make sense to augment it with some cloud resources to do that? We have lots of guidance on our website that will help walk you through that and we have solution architects that can help with the technical side of these deployments. Get going with that pilot group and then once those pilots are stable, we can look to roll it out beyond that.
Some smaller outfits or contractors may not have the resources to build out their own infrastructure, so Teradici partners with many different organizations to use this remote technology in different ways. We have a great partnership, for example, with a company called BeBop Technology. They provide Adobe Premiere Pro and other editorial products as a service, so that they will manage the infrastructure for you. On the bigger side, we work with Amazon WorkSpaces, which is one of the big Desktop-as-a-Service (Daas) providers, and that's all based on PCoIP. So, you can use Amazon WorkSpaces, you can use some of these more specialized ones like Tehama, BeBop or Avatara – these companies provide workstations as a service, so that's another way to go.
There are lots of case studies on our website, getting started guides, and ways to get you going.
Interested in learning more about how broadcasters can leverage Teradici PCoIP and Cloud Access Software to support remote work scenarios? Check out the information, case studies, and whitepapers we’ve collected on our page for Broadcast Studios.