Page 1


Virtualization and Radio

January 2018 From the Publishers of Radio World International


Virtualization and Radio

Sponsored by

January 2018 From the Publishers of Radio World International


Virtualization Technology Matures


Hot FM Moves Into Virtual Realm With Lawo


Don’t Fear Virtualization — Embrace It


Virtualization and Radio As a natural progression from the now widely adopted AoIP technology, an increasing number of media houses are beginning to explore the use of visualization in the broadcast facility. Still in its infancy, virtualization, or the dematerialization of hardware, appears to be positioned to offer real benefits to broadcasts. Obvious advantages include reduced hardware costs, space savings, improved effiMarguerite Clark ciency, easier updates and scalable architecture. But Content Director what about the potential challenges that may arise with the onset of this new technology? This latest Radio World International ebook examines the different aspects of virtualization in the broadcast realm and addresses such issues as station staff skill-set requirements; virtualization for mission-critical work; managing audio latency; VM server failure risks; licensing costs; and ways to ensure adequate redundancy. Radio World has published more than 50 ebooks exploring the many facets of radio station and network operations including mobile reporting, studio/transmitter links, visual radio, studio apps, social media, consoles, digital radio developments and more. Find recent issues at https://www.radioworld. com/resource-center/radioworld-ebooks. — Marguerite Clark

Seizing the Opportunities Virtualization Brings


Virtualization Adds Flexibility to Radio


Five Steps to Virtualization

VIRTUALIZATION AND RADIO Cover image: Thinkstock/metamorworks

Radio World International Edition | January 2019


Virtualization Technology Matures Will it revolutionize the broadcast facility or is it a solution in search of a problem? By Doug Irwin

being encapsulated from, the operating system where the application is located,” according to Techopedia.

Virtualization is an important, well-known and mature technology in the IT realm. By virtualization, I mean the creation of a virtual — rather than physical — version of a server, storage device or network resources. Operating systems can be virtualized as well. “Operating system virtualization is the use of software to allow a piece of hardware to run multiple operating system images at the same time,” according to WhatIs. A common use of virtualization is for servers. A software layer called a Hypervisor is used to emulate the underlying hardware — including CPU memory, I/O functions and network traffic. An operating system, which normally interacts with hardware, now interacts with a Hypervisor instead. “The main goal of virtualization is to manage workloads by radically transforming traditional computing to make it more scalable,” according to Techopedia. “However, deploying virtualization technology requires careful planning and skilled technical experts. Since the virtual machines use the same resources to run, it may lead to slow performance.” There are four types of virtualization that are germane to broadcast applications: server virtualization, as noted; storage virtualization; desktop virtualization; and application virtualization. Storage virtualization is the combining of physical storage from multiple network devices in to what looks like a single storage device. This is common in storage area networks. Desktop virtualization is straightforward. A user sits down to a monitor, mouse and keyboard, but the host computer that used to sit on top or under the desk is now virtualized. Communications from that desk back to the virtual server is done via a thin client (which is a terminal or a network workstation with a few resident programs that accesses most programs residing on a connected server). Application virtualization refers to the running an application on a thin client. “The thin client runs in an environment separate from, sometimes referred to as


I spoke with several industry people who have extensive virtualization experience during my research for these articles. I’ll start here by sharing my discussion with Mike Dosch, senior product manager, Radio OnAir, Lawo. “Virtualization could mean making software versions of products, i.e., the compass app on your phone replaces the thing with the magnetic needle you used to play with in the Boy Scouts — or it could mean the use of virtual machines and Hypervisors within an IT context,” said Dosch. “The compass app can run on a virtual machine, the physical compass cannot. We refer to the process of things becoming apps as ‘dematerialization,’ and it is an essential first step before we can talk about more exotic abstractions.” Dosch asks us to consider audio processing. “We still use boxes to perform this function and there is no benefit to virtualization in the case of a hardware processor,” he said. “It is a purpose-built device running in a proprietary platform. It must become software capable of running on a COTS [commercial off-the-shelf] platform before we can experience any of the benefits of virtualization. “Consider Stereo Tool — this product runs under Windows. It’s an example of a dematerialized radio program processor. It has most of the same features as the hard-

Mike Dosch

VIRTUALIZATION AND RADIO Radio World International Edition | January 2019


Thinkstock/Chad Baker

company mentioned by users, though there are others. Dosch continued. “Now that we’ve crossed over into software, we can begin to imagine the second type of virtualization: Rather than running my software audio processor on a Dell CPU tower computer, what if I ran it in a virtual machine (VM) running under a Hypervisor such as VMware on an industrial server running on multiple blades? In other words, what if I am running my software on an on-premises cloud?” He said this idea brings with it interesting benefits, mostly with the idea of scaling your infrastructure — building a second air chain, for example, is as simple as copying an environment — as well as redundancy. “Redundancy is especially interesting to me,” Dosch said. “The idea of the Hypervisor running multiple VMs is that it is aware of the status of both VMs and hardware and can take action in the event of a problem. A server blade fails? No problem. The Hypervisor runs your processes on a different blade and sends a notification to you about the server in need of attention. You stay on the air even though a catastrophic failure has occurred.” Dosch suggested that the broadcast plant will look more and more like a data center. A further implication

ware boxes and many say it performs on par with the very best of them, but since it is pure software running on a COTS computer under a COTS OS, it offers the benefit of a lower cost. Why is that? It is simple scale — computers are commodities and are manufactured in very large volumes. Purpose-built hardware made for radio is produced in much smaller quantity and will carry a correspondingly higher price,” Dosch said. “So here we see the first important benefit to virtualization: cost. The second might be ease of repair, upgrade or replacement. Again, since we are running this on a COTS platform, any activity requiring us to touch that platform will be cheaper than doing so with a proprietary device.” Dosch said some users might not be comfortable with this. “Some might prefer the idea of a box with ins and outs and it just does what it does and nothing more. That’s fine. If the only advantage was cost, I could imagine many would choose the simplicity of the box and eschew a computer-based product with its own unique annoyances — such as Windows security updates.” VMware is a well-known virtualization software provider based in Palo Alto, Calif. In my research VMware is the only

Continued on page 6 ❱

VIRTUALIZATION AND RADIO Radio World International Edition | January 2019


❱ Continued from page 5

studios, it could be justified. Or, maybe I use the public cloud in a new and interesting way — such as a backup source? Chances of a local failure and an AWS outage happening at the same exact time are quite small. “And I understand your point. This is very common with new technology,” he continued. “The first reaction is resistance because of this and that current failing. When we introduced Ethernet to replace audio routers, oh, my gosh! My favorite objection was: ‘I can’t even get my computer to connect reliably to the printer down the hall. You want me to put my audio on a computer network?’ And again, time and progress solves problems and the benefits don’t seem so risky.” He continued, “The idea of software to replace those

is that, perhaps, certain functions can be pushed out to “the cloud.” “If our listeners are using the internet to receive our streams, why can’t we originate our streams in AWS (Amazon Web Services) or other datacenter? We could run our processing and encoding in that cloud so we only need to send our unprocessed program signal to the datacenter where it is sweetened, encoded and replicated for our zillions of happy listeners,” he said. THOUGHTS FROM A SKEPTIC

I don’t want to portray myself as “too much of an old guy,” though when I started in the business, we were still using a vacuum tube console at one of our stations; the Orban 8000 was a new product. I’ve seen a lot of the evolution in broadcast technology in my 35+ years. Not surprisingly, I’m a skeptic when it comes to the application of this technology to radio studios. I wrote this back to Dosch: “Granted that day-to-day broadcast is not a life-or death situation — but I’m not sure I consider Hypervisor running VMware on two server blades to really be that redundant. How long does it take Hypervisor to note a problem and to move from one blade to the other? To what level can a “problem” be defined? Can I work on the issue while the other blade is operating? If they are physically located in the same container, can I work in that container while the system is hot?” Dosch responded that this is why he thinks it is so interesting. “These VM environments are designed for datacenters which are to have 100 percent uptime; a very wide range of redundancy is supported along with capacity scaling, alarming, load balancing, etc.,” he replied. “I agree with you, for one or two blades and one or two VM instances, we might not be exercising all of the capabilities. But once you turn everything into software, you begin to imagine things differently. For a single studio, I’d go with current tech. For a plant, I’d start to think about how this might help me.” I also told him that for me personally, the issue here is one of responsibility. “Once the infrastructure is moved to the cloud, which is really just a hyped name for a data center anyway, or a CO — telephones have been done this way since day one — I lose the ability to respond rapidly,” I noted. “If it weren’t for the well-publicized failings of AWS, I might have more confidence in the idea. I can’t really tell my PD or GM that the station is off the air till AWS fixes a network issue. Not going to fly.” Dosch replied, “One approach would be to exclude the datacenter for now — build an on-premises datacenter. It would not make a lot of sense for a single studio station, but for a plant with multiple stations, each with multiple

You’re talking about, for your typical radio station engineer, a whole different skills set ... You’re talking a hard-core IT guy. — A cautious industry observer devices with which we are comfortable makes us uncomfortable. I’m with you really ... but, software applications that replace all the purpose-built devices make sense for a number of reasons. It is disruptive and always disruption must be approached with caution.” I don’t think anyone would disagree that the use of a virtualized system of “routers” and associated control (done with thin clients) could be cheaper than the systems we are accustomed to in 2018. However, I would like to point out that new studio builds will still need space — often to accommodate a half-dozen people — along with associated furniture, lighting, air conditioning, audio monitoring (speakers and headphones), mic processing and the keyboard/mouse/monitor combo that uses the thin client to talk to the VM server(s). None of those things can be avoided as long as people are sitting in a studio talking on the radio. REALITY CHECK

So far we’ve painted a rosy picture of the idea of virtualization of audio systems. Now I would like to introduce another skeptical broadcaster in to the conversation. He will remain unnamed; suffice it to say that he’s a wellknown, major-market VP of engineering and IT with a dozen or so years of experience with virtualization. What makes him particularly valuable in this context is that he is well-versed in the normal “engineering” Continued on page 8 ❱

VIRTUALIZATION AND RADIO Radio World International Edition | January 2019


❱ Continued from page 6

aspects of radio as well. With respect to the idea of virtualization of audio control systems, he asked: “If audio control and routing transition to a software application, manufacturers will need to constantly update their code — not to add new features — but to maintain compatibility with the ever-updated OS. With this being such a niche product and not one with a ton of money at stake (i.e., radio), how do vendors intend to keep up cash flow and maintain their products?” He said he is not against the technical concept, and in fact suspects there are few that get as excited about it and experiment with it as much as he does; but he questions business case: How will this save operators money? Dosch responded: “That’s a fair comment, but it perhaps ignores the reality that all apps will eventually run on the same COTS hardware and virtualization hosts, there is really only one infrastructure to maintain. “For example, does the station run business, playout, scheduling, traffic apps? Those can certainly run in the virtual environment. So stations will already be dealing with this in some sense. Extending the infrastructure capacity to also include a few more apps might not be as burdensome as you imagine.” I asked our skeptical engineer specifically how he would feel about using virtualization for an audio system for broadcast. “What makes me nervous,” he replied, “is that, in doing audio the way we currently do it, it’s very sensitive to latency. The hardware you would need to handle this would be beefy,” he said. “I would just be really wary of stuff that I consider mission critical, personally. I’ve been experimenting with virtualized work stations for a good amount of time and unless you have some very big hardware and a very fast storage network, there could be problems. Because, remember, the disks are shared. You can get some lag, as some machines won’t respond quite as fast. “There are a lot of aspects to virtualization that need to be managed,” he said. “You’re talking about, for your typical radio station engineer, a whole different skills set than what they’re used to. You’re talking a hardcore IT guy — somebody that understands virtualization, which is one specialty; somebody that understands storage, which is another specialty. So I get a little concerned. When this stuff works, it works great. But when it stops working, finding the problem can be very difficult and time consuming and that’s not something we have a lot of in the (radio station) environment.” But for any technology advancement to work, said Dosch, requires a combination of innovation and pragmatism to achieve the desired results. “Think of it this way: Every objection raised is a problem to be solved. And if we solve those problems, we make a better product and the technology advances. It always works this way.”


Wheatstone Systems Engineer Kelly Parker thinks virtualization technology will be applied in broadcast studio applications sooner rather than later. “We’re already putting processing, routing, mixing — all of that — in to the software realm, and cloud-based is the next step. There are a number of obstacles that we as an industry face, like real time monitoring. In the ‘cloud’ environment, there’s more latency. You press an ‘on’ button on a screen or a surface or whatever it might be, and there’s a reaction time that takes place. It takes time for that information to reach the data center, start the playback or automation, and to get it mixed and then back down to the operator so he can hear what he’s doing. That’s a big challenge. But we’re exploring all kinds of options for the future that will be without hardware. It’s where it’s all going.” I asked Parker about his thoughts on when this kind Kelly Parker of transition will start in our field. “I still think that there’s enough people out there that like that tactile feel of a fader and switch, so I don’t think it’s going to happen right away. I think people will accept it over time, but I don’t think it’s going to happen tomorrow,” he said. “The industry is pushing it — we’re working with clients on — virtualizing more and more of the operation.” When I asked about how this change would affect the day-to-day work of a broadcast engineer, Parker said to expect change, just as our skeptical engineer did. “Well, it’s going to change the structure of engineering. There may not even be studio guys, of which there are fewer and fewer of these days anyway. It’ll be more IT centric and probably collocated in some data center some place in the country, or several which back up, so I see that being a challenge, getting the people in to manage that. It will certainly take away from traditional broadcast engineers.” THE VIRTUALIZED BEEB

I also communicated with Kirk Harnack of the Telos Alliance about virtualization. “The BBC has been virtualizing its BBC Local Radio studio back-end operations since 2015 in a project called ‘ViLoR,’ or Virtualized Local Radio,” wrote Harnack. “Now, nearly all of all BBC Local Radio stations are using their familiar local studios, but the playout systems, phone hybrid pool, audio codec pool, and even the live console

VIRTUALIZATION AND RADIO Radio World International Edition | January 2019


mixing engines are located in redundant data centers. The Telos Alliance and Broadcast Bionics developed and provided all of the virtualized mixing consoles and talk show systems. “Some users of Axia and other Telos equipment have worked with us on proof-of-concept projects where playout, mixing, and phones are all virtualized in a VMWare environment. Continuing on the success of these proofs and the BBC’s proven ViLoR infrastructure, the Telos Alliance is developing systems specifically for ‘cloudifying’ audio acquisition, transfer, routing, mixing, and audio processing. Users will have options to use these solutions locally, in their own data center, or with publicly-available cloud computing strategies,” Harnack wrote. While console/router manufacturers are now discussing the virtualization of their systems in the not-too-distant future, at least one playout system manufacturer has a product on the market. ENCO Systems says that its ENCO1 is a “fault-tolerant, virtualized solution designed around a unique specification for the radio automation environment.” The traditional automation infrastructure places a workstation in each studio, and that physical box connects to a central server, or operates as part of a separate production or on-air cluster that shares a common network. Inside of a single ENCO1 “box” are two redundant sets of components: Each represents a complete server, and the hardware is mirrored between them so that any failure of any single component does not take the system down. The entire environment — all the storage, all the databases, all of the play lists and every bit of the workstation environments are preserved within this box,

which is installed in the protected environment of a station’s technical core. The traditional architecture can also require more maintenance involving the studio workstations because many of these boxes are different from one another, using different motherboards and hard drives that sometimes require replacement. With ENCO1 , the remote “workstations” are virtualized in the server itself. In each studio lives a small appliance that is used to connect to the server (via Kirk Harnack Ethernet). It provides the user with the mouse, keyboard and touchscreen control to operate the software. Up to 40 virtualized workstations can be supported. OPPORTUNITY

There is certainly plenty of opportunity in the global market to migrate broadcasters to virtualized playout. “Whether a large network or a small cluster of FM and/ or AM stations, most operations have at least several on-air and production studios with dedicated equipment,” said David Turner, executive vice president of ENCO. “This means a collection of space-consuming hardware distributed throughout the facility, often operating autonomously.” In an ENCO1 deployment, the use of the virtualized server and remote clients also frees up valuable rack space all while saving on electricity and the generation of audible noise. The thin clients operate as compact and generic computing devices that boot and run remote sessions, which are displayed on the studio monitor. “The hardware itself has a component that can phone home and tell the factory if there’s a problem with any piece of the hardware,” said Turner. “Then the factory can react and send a replacement component out to the station. It’s done in such a way that you don’t require a highly skilled operator or technician to be able to do it. It’s generally a replacement of a hard drive or a module within the system.” The evolution of technology in broadcast engineering is on-going and inevitable. The trend towards virtualization will, at least eventually, take hold. “Big groups all over the world are driving this,” said Wheatstone’s Parker. “It’s not just the iHearts of the world — it’s Media Works, it’s Penumbra in Australia, it’s RTÉ in Ireland, and the BBC. Everybody’s pushing to change the way we do things and it’s going to happen.” Harnack agreed. “It’s absolutely inevitable that broadcast operations will move to a virtualized, “cloudified” paradigm. There will be a few bumps along the way, but in a few years, we’ll wonder how we did it any other way.” n

David Turner of ENCO demonstrates the ENCO1.

VIRTUALIZATION AND RADIO Radio World International Edition | January 2019


Hot FM Moves Into Virtual Realm With Lawo Zambian station uses crystalCLEAR and R LAY to heat up the country’s airwaves

Station owner Oscar Chavula is pictured in the middle with station talent in the Hot FM studio.

By Marguerite Clark

ment and very little setup time,” he said. “We are the first in Zambia to do this and the innovation is here now.” The crystalCLEAR virtual mixing console features a software-based control surface and is driven by a multitouch interface on a high-resolution computer display controlling the Lawo Compact Engine. R LAY software implements computer virtualization technology adapted from the IT industry, letting broadcasters mix, route and process AES67 audio streams and computer audio via standard WDM or ASIO interfaces. “HotFM is one of the most modern radio stations in Africa — that’s why they chose Lawo,” said Meck Phiri of Lawo’s Zambian sales partner, Meck Media Consult, noting that Lawo radio products are AES67-compliant. “HotFM will deploy Visual Radio shortly; Lawo’s standards-based AoIP infrastructure makes that possible.” n

Hot FM in the Republic of Zambia is using Lawo’s crystalClear and R LAY virtual mixing radio consoles in its new studio that went on the air Aug. 1. The commercial station, based in the country’s capital and largest city Lusaka, with almost 2 million inhabitants, is active with numerous on-location live broadcasts, news coverage and interaction with listeners via social media platforms. It covers the region on three FM frequencies. Hot FM owner, Oscar Chavula, stated in a press release that touchscreen mixing brings with it many advantages because it’s easy to learn and use. “In the field, a computer with R LAY and a 4G telephone connection makes it possible to originate high-quality remote broadcasts with much less equip-

VIRTUALIZATION AND RADIO Radio World International Edition | January 2019


Don’t Fear Virtualization — Embrace It There are many paths toward virtualizing broadcast infrastructure


replaced. And today’s PC-based playout systems are even better than the CD players — and their audio skipping — of the 1980s and 90s. Over the past 20-plus years, we’ve “virtualized” the operation of many devices that were previously “purpose-built.” And while PCs proliferated throughout broadcast facilities, some vexing problems attended the PC takeover. The life cycles of hard drives, cooling fans and power Kirk Harnack supplies conspire to bring regular failures to any facility that depends upon dozens of PCs for continuous operation. Seems we’ve replaced purpose-built audio equipment with PCs performing specific duties, and our device count may not have changed much from our older scenarios. Tech businesses in other sectors have had similar challenges. Many of these challenges have been addressed using far more robust computing platforms than PCs. “Server-grade” hardware — including motherboards, CPUs, cooling fans, hard drives and power supplies — provide a level of performance and reliability not found in PC-grade systems. But server-grade hardware is expensive — too expensive for many broadcasters to replace their consumer PCs with a similar number of server-grade computers. This price point is where computer virtualization raises its hand to say, “Hey, I have your solution right here!” That solution involves virtualizing those very systems even further by moving their functions out of numerous dedicated PCs and into a server-grade platform, plus a full backup. These servers run multiple virtual “PCs” internally — anywhere from two or three systems up to several dozen. Some broadcasters began shifting their internal business processes onto multi-platform servers as far back as 10 years ago. But moving real-time audio processes to virtualization servers is just now gaining interest. One early experimenter — and then early adopter —

By Kirk A. Harnack The author is senior solutions consultant for The Telos Alliance. Most of radio’s operational systems can now be “virtualized” or soon will be capable of it. The implications for broadcasters will be either minimal or staggering, depending on each broadcaster’s business and operational models in the future. The transition to virtualization — the first form of it — actually began in the late 1980s with simple PC-based automation systems replacing tape cart machines. Long gone are squeaky pinch rollers, worn tape heads and lubricated tape sliding against itself in endless loops. And while the transition to all-digital storage and playback took nearly two decades, we learned a lot along the way.

Just as other industries have demanded, and now use, better computing hardware and more reliable operating systems, radio broadcasting benefits from the same improving technology as other business-tech sectors. TRANSITION

Are computers perfect replacements for the purpose-built equipment we’ve come to romanticize? Hardly. Yet despite their shortcomings, PCs have proven to offer better overall usability, reliability and audio quality than the cart machines, reel-to-reel decks and turntables they

VIRTUALIZATION AND RADIO Radio World International Edition | January 2019


has been BBC Local Radio. The BBC operates “local” radio stations in about 40 cities across England. EXPANSION

In 2014 the BBC embarked on a quest to reduce the difficulties and expenses associated with local equipment capital and maintenance costs, as well as a shortage of engineers to refit 80 complex studios. The result is “ViLoR,” short for Virtual Local Radio. The local studios operate about the same as with the old, legacy equipment, but the underlying broadcast Radio control room virtualization using IP-Tablet software. infrastructure lives within two data centers in London and Birmingham. tive multi-platform distribution. Automation playout, talkshow systems, audio codecs Recent developments are bringing broadcast infraand even live on-air audio mixing are handled at the data structure virtualization to more broadcasters — even centers by high-reliability servers. Redundant data conthose without large budgets. Since “virtualization” nections and dual studios / data centers assure six 9’s of embodies several definitions and manifestations, broaduptime. casters can virtualize a few things now, then move The BBC is using resources more efficiently than before toward radical virtualization of their infrastructure when in terms of capital costs, maintenance and monthly the time is right. expenses such as HVAC. The new ViLoR system maintains A contemporary example of virtualization is moving audio in digital perfection with extreme ease-of-use the audio console surface plus phone interaction and for the on-air talent. Serendipitously, ViLoR affords a renewed emphasis on talent, content creation and effecContinued on page 14 ❱

Multiple radio automation systems operating in a robust and redundant VM server.

VIRTUALIZATION AND RADIO Radio World International Edition | January 2019


Conceptual connectivity and functions involved in a fully cloud-based radio broadcast infrastructure. ❱ Continued from page 13

of new audio isn’t duplicated by each station; rather it’s done once by another VM running an automated ingest and quality-control application called WebGopher. This application has freed up our local talent from the tedium of manually dubbing the hundreds of audio cuts that are updated daily. Now they can focus on their shows, voice-tracking and doing frequent remote broadcasts from local events and advertisers. A further virtualization scenario appears when we replace audio mixing hardware (dedicated console “engines”) with audio mixing engines running on PCs — or better yet, on VMs. Audio over IP technology is making this quite doable. As mentioned, the BBC began such operation nearly five years ago. When mixing engines are virtualized, radio studios may have simple audio mixing control surfaces, along with mics, headphones, touchscreens and button panels; the heavy lifting of the rest of our typical infrastructure will all be virtualized in clean, safe data centers, either on-site or in a secure off-site facility. Webcasters are already operating some of their facilities entirely in the cloud. The features and functions that broadcasters need are being developed now. Will virtualized broadcast infrastructure be better, more reliable, cost less? Will it allow talent to focus more on their craft — and do that from anywhere? Stay tuned. The future does look virtual, and that’s a reality we’ll have to deal with. n

other studio controls to a touchscreen. Today’s multitouch screens offer effortless and accurate control, and the layouts are customizable for each show or on-air talent. They can be less expensive than the traditional controls they replace. Another form of virtualization is to move automation playout systems onto Virtual Machines (“VMs”) running on powerful, server-grade platforms. Some automation manufacturers are supporting this scenario already, and most are planning on it. At Delta Radio in Mississippi, hardened servers are running five instances of the Rivendell automation system. Audio and GPIO signalling is done entirely in Livewire Audio over IP, so no sound cards or GPIO cards are needed at the server. The server offers four 1 Gb Ethernet connections, of which only two are used — one for the Livewire AoIP network and the other for VNC access, audio and log file transfers, and other command/control functions. Once set up and tested, these automation systems have worked flawlessly. Two of the stations are entirely automated while three have some local, live-assist shifts. All of these stations ingest voice-tracking files daily, as well as some long-form programs, hourly weather forecasts and daily local news files. Actually, the automation systems running in VMs all share a Network Attached Storage (“NAS”) server. Ingest and formatting

VIRTUALIZATION AND RADIO Radio World International Edition | January 2019


Seizing the Opportunities Virtualization Brings Broadcasters must leverage the almost limitless flexibility, scalability and efficiency available Dan McQuillin

GUESTCOMMENTARY By Dan McQuillin The author is managing director for Broadcast Bionics. Three overlapping trends are leading us toward an increasingly virtualized future, transforming the design of our broadcast studios and facilities over the next decade as our traditional hardware and tools are replaced by virtual software interfaces, running on generic virtual machines (rather than dedicated hardware), increasingly deployed and managed remotely in a public or private cloud. To those looking to succeed in revolutionizing their content and business models, this is not just a matter of emulating or replacing our existing user experiences, workflows and facilities, but seizing the opportunities virtualization affords to enhance them radically and leverage the almost limitless flexibility, scalability and efficiency available.

already mostly been replaced by computer hardware, albeit embedded into discrete, dedicated appliances. Liberating all this software masquerading as hardware into generic virtual machines allows organizations to leverage the scalability, management and resilience tools developed to manage modern data centers to deliver much more efficient broadcast architectures and facilities.

Three overlapping trends are leading us toward an increasingly virtualized future.


Once our interfaces are software and the architecture virtualized, we will increasingly see local systems move to dedicated data centers as part of either a private or public cloud. The capital equipment, real estate and running cost savings available make this an economic imperative for broadcasters business models. There are huge savings to be made using generic virtualization technologies. We should however take care that we reap the wider benefits available, not just to deliver our existing workflows and services more efficiently but to enable new experiences and platforms that will serve and delight audiences long into the future. The pace at which services and platforms change demands a much greater flexibility for the way our facilities are architected. Studios that once delivered a

A few years back replacing your console with a piece of glass felt like a cultural impossibility. With talent entering our industry from the iPad and YouTube generation, our traditional technology no longer looks inevitable but increasingly archaic.  We need to encourage new talent, new workflows and new business models for broadcasting to survive; this requires much more radical and flexible thinking about the production spaces we build and the technology we make available. Reducing the specialization and skills required is not just about reducing cost, it also democratizes access to production and creates more flexible content that is fit personalized experiences and social platforms of the future. The dedicated broadcast hardware and DSP once required to reliably deliver high-quality content has

Continued on page 19 ❱

VIRTUALIZATION AND RADIO Radio World International Edition | January 2019


Virtualization Adds Flexibility to Radio A new approach to simplify workflows and enable mobility On-Hertz Founders Renaud Schoonbroodt and Benjamin Lardinoit

GUESTCOMMENTARY by Benjamin Lardinoit The author is cofounder of On-Hertz. In a world in which consumers have almost unlimited choice about the content they consume and where and when they consume it, how do you create loyalty? How do you make your content “sticky” so that those consumers keep coming back for more? Offering tailored content for an audience is one way. Increasing interaction is another. Being relevant and topical is yet another. Creating a sense of community, of belonging, is a fourth. In the face of these challenges, and in an environment where on-demand TV and social networks are well established, a new medium is arising. It’s one that’s more widely available than 4G. It excels at being “live” — but can also be on-demand. It works for every type of consumer. It enables those who use it to explore topics of interest in greater depth — or to discover new areas of interest. It connects people, engendering a sense of community, a sense of belonging — and it does that with no risk to anyone’s personal data. It sounds like everything you could ever want, right? But you’ve already got it, and as you’ve probably guessed it’s called radio. Radio has long played a central part in our lives — and it can continue to do so if it plays to its strengths of immediacy, of responsiveness, of relevance, of its ability to create conversation and to share opinions and experiences. If it delivers on those, it can create a depth of engagement and loyalty that other media formats can only dream of.

get in the way. Setting up a radio studio in a muddy field and coordinating the logistics and resources necessary is arduous, challenging and costly. Too often, those considerations deter radio broadcasters from “getting out there” and creating that sense of excitement, spontaneity, intimacy and, yes, fun that is only possible from being on location, close to the action. Of course radio has covered music festivals for many years. Big events like that are worth spending time, money and effort on — and research proves that they generate an audience response that makes it wholly worthwhile. But what if it was possible to cover much smaller events, and cover them more frequently without the time, money and effort? Or what if you thought it might be fun to broadcast a show from a hot air balloon — in full sight of your audience? Or from the top of a mountain? Or at a local village fair? Or even from home? What if your imagination could run wild, unrestricted by considerations of logistics, upheav-


To do so, though, mobility is key. Imagine, as a producer, being given the choice of covering a music festival from the studio or from the event itself. It is, as they say, a no-brainer. Or it would be if logistics and cost didn’t

VIRTUALIZATION AND RADIO Radio World International Edition | January 2019


on the one hand, or accepting a reduced level of functionality and capability on the other. Virtualized radio means that you can, in effect, have it all: minimal logistical effort with maximum quality of output. But it’s because of its impact on the bottom line that virtualized radio is really getting attention — because it can increase revenues while decreasing cost. Taking the show on the road gets it closer to its listeners. It makes content more relevant to their interests. Those two characteristics mean boosted audience reach and engagement and make the show much more attractive to advertisers.

al and expense? From where might you broadcast then? And here’s another thing. You’d like to think that technology has made our lives simpler. In radio, however, it really hasn’t. It’s allowed us to do more, but it hasn’t made broadcasting a radio show any easier. It’s as if the Apple revolution never happened — a revolution that meant that anyone could use, and get value from, any device because it was so simple to master. Which is a shame because, at its heart, radio couldn’t be simpler. It’s people talking into microphones, playing songs and interacting with each other. It is now time to bring radio production workflows to the same level we demand from all our other devices. Why? Simply, because in radio as in all other forms of media, content is king — and every minute and cent we spend on something that isn’t content is time and money wasted. A radio presenter’s value doesn’t lie in his or her ability to watch dials, twiddle knobs, move faders. It lies in being able to deliver content in a way that’s entertaining and engaging.

In a virtualized radio world, broadcasters can produce shows more easily because less effort is required, thus more shows become a reality.


Then there’s the issue of getting the most value from scarce and valuable resources. You’ve set up that outside broadcast from that music festival — so is it the best possible idea to chain your technical team to the remote studio when their work is, to all intents and purposes, done? Of course it’s not. All of the above is leading us to the concept of virtualized radio, which has at its heart the goals of simplifying workflows and reducing cost while at the same time enabling an unprecedented mobility that can allow radio to do what it has always done best — and must continue to do if it is to continue to thrive. Unsurprisingly, the concept is capturing significant interest. Today, an outside broadcast means making a choice between undertaking a major logistical exercise ❱ Continued from page 15

single dedicated service for decades are now required to be repurposed and reengineered on a daily or even hourly basis. Entire platforms, services, monetization and content opportunities now come and go in a matter of months. The future will be won by the agile. TIPPING POINT

While this virtual future will not happen overnight, we have reached a tipping point. Many of the trends and technologies that will deliver the transition have been accelerating in development and deployment for some time. Broadcasters should already be looking for and investing in solutions and suppliers that are delivering software and systems that can run on generic virtual machines with interfaces delivered over html5 to any

In a virtualized radio world, broadcasters can produce shows more easily because less effort is required, thus more shows become a reality. The threshold that makes a show viable becomes significantly lower. In addition, virtualized radio is inherently much less expensive to own and operate than today’s more cumbersome approach. That means more money to invest in content. Trucks don’t have to roll or crews deployed, not to mention reduced replacement and maintenance cost as well as no need for studio space. It’s no surprise therefore that virtualized radio is something that many broadcasters are evaluating actively. The new approach of virtualized radio offers vast potential to an old but reliable medium. n browser. The public cloud does not yet quite deliver the latency and predictable uptime required for broadcast critical systems, but local data centers and private clouds are already part of the broadcast chain for many major broadcasters and the migration is underway. I have had the privilege of working with some of the world’s leading broadcasters and equipment providers to prove the technology is reliable and demonstrate that compelling benefits are achievable.  The projects we have been involved with have demonstrated significant practical and economic rewards await those willing and able to embrace virtualization.   The building blocks are in place. We must embrace this change. The future of our industry, creatively and economically, will be built upon the flexibility and functionality now available to us. What are you waiting for? n

VIRTUALIZATION AND RADIO Radio World International Edition | January 2019


Five Steps to Virtualization What this new technology means to the broadcast operation Dee McVicker


From an AoIP manufacturer’s point of view, I can tell you that the broadcast studio is further along than one might expect. More and more, hardware is being morphed into software apps. We now have mixer GUIs up on tablets and, in the case of Wheatstone’s AoIP system, virtual mixers inside the network I/O units themselves. EQ dynamics and compression, as well as signal monitoring and control, are on tap throughout the AoIP network. We have complete remote control over devices, workflows, and signal flow in the station studio and at the transmitter site — and outside the studio, too.

By Dee McVicker The author is associated with Wheatstone Corp., which manufactures the WheatNet-IP audio network. Someone once said that if you can walk with your head in the clouds and your feet on the ground, you can become a professional basketball player. That’s also good advice for broadcasters today. Broadcasters would do well to keep their feet on the ground as they begin to think about what virtualization means to the broadcast operation, most especially this concept known as “radio in the cloud.” Most of the technologies for a cloud-based radio operation exist today. Corporate enterprises are making


In truth, the cloud has already started to form in the studio and with this, we’re seeing real operational cost savings for broadcasters. Every station’s path to cloud and virtualization will be different, but we can break them into five basic steps:

More and more, hardware is being morphed into software apps.

Virtualizing the functions and control of local hardware. This step is taking place already with the development of virtual mixers and the use of virtual development tools. We can control more functions from one tablet and an IP connection than we could ever dream of controlling from a bank of hardware. In short, we’ve mobilized mixers, silence detection, routers, audio processing — all the software that already lives in the IP audio network.


good use of Software as a Service to centralize and scale their operations, and the supervisory technology is fairly mature for overseeing deployment of these resources. Meanwhile, private and public cloud service providers are doing brisk business. Two of the biggest providers, Amazon Web Services and Microsoft Azure, are at least doubling in business year after year. All of this will be important for consolidating stations and workflows. That is the 35,000-foot view of a cloud-based operation. But what is happening with radio at the ground level?

Virtualizing the boxes. Next, we’re steadily moving all the functions of existing boxes to apps that can run on a computer. Since most of our functionality is already in software, we’re not necessarily starting from scratch.


VIRTUALIZATION AND RADIO Radio World International Edition | January 2019


INTERNATIONAL EDITION Move all the local computers to VMs on a local server. Moving computing to local VMs, or Virtual Machines, in the studio building is likely to be the next step. This will allow broadcasters to maintain real-time program generation in the studio, without the latency and bandwidth availability issues that currently make off-site cloud operations difficult to maintain over an internet connection. VMs create the desktop environment in software for running various applications used in the studio. Thin clients, which could be a strippeddown application device or may even be an app on your laptop, access the apps on the VM, thereby simulating a “cloud” model. Automation companies are already centralizing automation on virtual machines, and manufacturers like Wheatstone are developing ways to increase the redundancy of these systems, similar to what was done with distributed AoIP networking.


Move the VMs from a local server to multiple server farms somewhere else, such as regional studios or colocation facilities. This is one type of cloud. This model requires cloud supervisory technology that manages access to and maintenance of those machines in a way that keeps track of what they are doing. Think of it as a local/virtual switch — sort of — in the cloud for making sure all the VMs interact the way they would as desktop machines, or in the case of WheatNet-IP, I/O BLADEs on your local network.


Move operations to a Cloud Service Provider (CSP). The advantage of cloud service providers like Amazon’s AWS and Microsoft Azure is that they can run virtual machines in several data centers simultaneously, with automatic load balancing and latency monitoring automatically routing users to the closest, fastest data center resource. This is an important cloud principle. When you put something on Dropbox, you’re not putting it on one server. You’re putting it on many servers strategically located all over the country. When you request it again, wherever you are, the server you’ve got best access to will automatically serve you the information. That is a true cloud.


A few challenges remain for the station in the cloud model. For example, latency is a huge issue for real-time content and for controlling devices. But as is true of all IT-related matters, it is just a matter of time before broadcasters adopt cloud technologies in their own unique way and for their own unique purposes. n

Email with “Letter to the Editor” in the subject field.

January 2019 FOLLOW US

CONTENT Managing Director, Content Paul J. McLane, Content Director Marguerite Clark, Senior Content Producer — Technology Brett Moss, Content Manager Emily Reigart, Technical Advisers Thomas R. McGinley, Doug Irwin Technical Editor, RWEE W.C. “Cris” Alexander Contributors: Africa: Gregory Lagat, Lameck Masina, John Masuku; Asia: Md. Azizul Alam Al-Amin, Martyn Green, Frederick Noronha, Mark Timpany; Australia: James Cridland; Europe: Andy Bantock, Kevin Branigan, Giuseppe Camonita, Nuno de Sousa, Charles Flores, Blazo Guzina, Lawrie Hallett, Will Jackson, Hans K. Kristjansson, Drew Leifheit, Raúl Llarull, Urmas Loit, Marc Maes, Davide Moro, Emmanuelle Pautler, Panos Polyzoidis, Bernd Trutenau, Andreas Tzanakos, Thomas Völkner; Latin America: Jorge J. Basilago, Carlos Eduardo Behrensdorf, Juan G. Buenaventura; Eduardo Curuchet, Arturo Valentino; New Zealand: Nick Gerritsen; North America: John Bisset, James Careless, Scott Fybush, Randy J. Stine Production Manager Caroline Freeland Managing Design Director Nicole Cobban Senior Design Director Karen Lee ADVERTISING SALES Publisher, Radio World International Raffaella Calabrese,, +39-320-891-1938 Senior Business Director & Publisher, Radio World John Casey,, +1-212-378-0400 x512 LICENSING/REPRINTS/PERMISSIONS Radio World is available for licensing. Contact the Licensing team to discuss partnership opportunities. Head of Print Licensing Rachel Shaw MANAGEMENT Managing Director/Senior Vice President Christine Shaw Chief Revenue Officer Luke Edson Chief Content Officer Joe Territo Managing Director/Europe Mark Burton Head of Production US & UK Mark Constance VP/Media Technology Group Carmel King FUTURE US, INC. 11 West 42nd Street, 15th Floor, New York, NY 10036-8002

All contents © 2018 Future Publishing Limited or published under licence. All rights reserved. No part of this magazine may be used, stored, transmitted or reproduced in any way without the prior written permission of the publisher. Future Publishing Limited (company number 02008885) is registered in England and Wales. Registered office: Quay House, The Ambury, Bath BA1 1UA. All information contained in this publication is for information only and is, as far as we are aware, correct at the time of going to press. Future cannot accept any responsibility for errors or inaccuracies in such information. You are advised to contact manufacturers and retailers directly with regard to the price of products/services referred to in this publication. Apps and websites mentioned in this publication are not under our control. We are not responsible for their contents or any other changes or updates to them. This magazine is fully independent and not affiliated in any way with the companies mentioned herein. If you submit material to us, you warrant that you own the material and/or have the necessary rights/permissions to supply the material and you automatically grant Future and its licensees a licence to publish your submission in whole or in part in any/ all issues and/or editions of publications, in any format published worldwide and on associated websites, social media channels and associated products. Any material you submit is sent at your own risk and, although every care is taken, neither Future nor its employees, agents, subcontractors or licensees shall be liable for loss or damage. We assume all unsolicited material is for publication unless otherwise stated, and reserve the right to edit, amend, adapt all submissions.



Write to RWI

VIRTUALIZATION AND RADIO Radio World International Edition | January 2019


Future plc is a public company quoted on the London Stock Exchange (symbol: FUTR)

Chief executive Zillah Byng-Thorne Non-executive chairman Richard Huntingford Chief financial officer Penny Ladkin-Brand Tel +44 (0)1225 442 244

Profile for Future PLC

Radio World International - Virtualization and Radio - January 2019  

Radio World International - Virtualization and Radio - January 2019

Radio World International - Virtualization and Radio - January 2019  

Radio World International - Virtualization and Radio - January 2019