Top Ten Telecom Tips
Anthony Harrington takes a look at developments in telecoms technology
Anthony Harrington takes a look at developments in telecoms technology
Five years ago virtually all the “issues” raised here would have been predicated on either laboratory, blue sky or barely emerging technologies. Other issues would have dominated. Deregulation of the sector would have figured way up the list, as would ISDN and leased line charges. Things, however, have moved on. All may not yet be perfect in telco land but there is a sense that everything is up for grabs. New entrants, using bleeding edge technologies, are rewriting the rule books and reshaping business models. Competition is the name of the game and telcos are displaying an urgency and energy that would have astonished telco executives of an earlier era
Unified Networks: dealing in packets
One of the more highly publicised transformations of the sector is the increasing dominance of Internet Protocol, or IP, as the technology of choice for the future as far as both voice and data is concerned. In the corporate environment, this opens the way for enterprises to simplify their networking and communications infrastructures. Instead of running separate voice telephony and data networks, the idea of using data networks to run Voice over IP services, is gaining ground. Running a unified network is a simplification exercise and holds out the possibility of efficiency gains and of more tightly integrating voice and data applications across the company and between the company and its suppliers and customers.
It also holds out the possibility of voice calls “piggy backing” for free on existing corporate data networks (whether leased line or virtual).
However, many analysts and industry figures have warned that the so-called “toll avoidance” argument is not a sound one, since competition is forcing down the price of international and local telephone calls. Instead, they say, companies interested in exploring a unified networks strategy should look for their ROI from efficiency gains and the benefits of new, tightly integrated applications.
For telcos, however, the strength of demand for IP services has led to a major strategic rethink over which protocols hold the key to the future of broadband services for corporates. Five years or so ago, as Peter Boland, vice president of product strategy at Nortel Networks points out, things were very different. ATM, a cell-based connection orientated protocol with strong management and quality of service features, had become the technology of choice for telco networks and was being adopted by corporates as a backbone technology.
At the time there was talk of ATM providing a common LAN and WAN infrastructure.
In fact Ethernet won the battle with ATM for the corporate LAN, and while IP can be layered on top of ATM, many telcos now increasingly view ATM as an unnecessary layer in the network.
Chris Lee, managing director for the WAN Systems Group at Lucent Technologies, has a crisp way of putting this: “In the future all we will have is light and IP.” The point here is that as carriers roll out massive amounts of IP bandwidth, all traffic, voice and data, turns into IP packets, so networks in the WAN space are unifying anyway. This in turn will help to push corporates more naturally into investing in unified networks as they seek to take advantage of next generation IP services provided by telcos, Internet Service Providers (ISPs) and Application Service Providers (ASPs).
The death of distance: the end of easy margin
The traditional formula for pricing services in the PSTN world of switched circuits was based on a multiple of time and distance. This was always a somewhat artificial formula, complicated by complex interconnect agreements among carriers on international routes. In the new world of optical switches, using dense wave division multiplexing technology (DWDM), the idea of relating charging to distance loses all credibility. The add on costs for distance, factored over the sheer volume of traffic that optic fibre can transport, tend towards zero. Instead, under the pressures of intense price competition, wholly new cost models are being experimented with (a fact that has led to this phenomenon being renamed as “the death of margin”).
Bob Cushing, group marketing and business development director at Redstone Telecom, comments: “The idea of basing my charging on the number of minutes of voice or data that I switch through my network is dead. Basic services are now provisioned on a monthly rental, with high speed Internet access provided as an always on service with no supplementary call charges.”
Some bleeding edge telcos are already exploring the potential of charging corporate clients on a “per seat” basis for a fixed amount of bandwidth, a model that treats communications much like any other application.
Optical switches: how fast is fast?
Today’s optical switches are capable of delivering throughput rates at 10 gigabits per wavelength, with some vendors either already shipping or planning to ship 40 gigabits per wavelength. Since DWDM technology is now able to separate out 160 wavelengths inside a single fibre, this adds up to 160 x 40 gigabits of raw throughput per fibre. This is an enormous amount of capacity in anyone’s language. Another multiple worth adding to the equation is the fact that some cable suppliers are now laying cable with 600 fibre pairs, a “pipe” with the potential throughput in today’s terms of 7,680,000 gigabits, (7,680 terabits, or 7.7 petabits).
Keith Gibson, sales and marketing director at Fujitsu Telecom, emphasises that this is just the beginning for the industry. The theoretical upper limit for optical switch technology is the speed of light. So far the industry is only able to pulse signals onto a wavelength at a modest fraction of the speed of light. By way of contrast, silicon chip technologies are already at 60 percent of the medium’s theoretical limit.
This means that it requires much less capital investment to go faster in optics right now than it does in silicon, so further breakthroughs are inevitable. It might be hard for many people to realise, as they are limited to today’s dial up modems operating at 56.6Kbps, but the fact of the matter is that bandwidth abundance is truly just around the corner.
This brings us back once again to the fact that simply selling high speed data carriage (with voice being just another data type), is likely to be a losing proposition except on the grandest scale.
Dynamic bandwidth provisioning: as much as you want, whenever you want
While optical networks are making huge amounts of bandwidth possible, perhaps the most exciting new development for many telcos is the prospect of managing this bandwidth dynamically. The idea is to allow customers to turn bandwidth up or down dynamically, in real time as need dictates.
Andy Wood, chief technology officer at Storm Telecommunications, for example, reckons that dynamic bandwidth management is at the heart of a new breed of value added services the company plans to offer its clients.
Alan Taylor, consulting engineer for EMEA at Juniper Networks, looks forward to a future where terabit speed IP routers such as Juniper’s M60, can automatically provision additional bandwidth from various carriers’ optical networks in accordance with pre-set policies. Today, provisioning a new circuit for a client takes a telco considerable effort since it has to configure that circuit route throughout its network. Instant bandwidth provisioning that ripples through automatically into billing will be a huge advance in the “value add” that the sector can offer organisations.
Before this can happen, the industry has a substantial amount of work to do agreeing on the hooks that will allow one carrier’s network to automatically provision bandwidth from several other carriers’ networks. Already, however, consortia are at work to do just this. The Optical Domain Service Interconnect (ODSI) consortium has a goal of writing a common protocol, by the start of 2001, that will enable exactly this kind of cross-vendor inter-operability.
CRM: helping to track the customer
Both deregulation of the telecoms sector and the rise of e-commerce have highlighted the crucial importance of finding, retaining and “farming” profitable customers. This has made telcos one of the biggest markets for systems integrators selling customer relationship management (CRM) systems. Neil Stevenson, lead telco strategist in IBM’s Global Services division, begins from the premise that the number of telcos in the world who will be able to compete on a pure price model is small and diminishing.
For the vast bulk of providers, the key survival strategy, he notes, will involve a services strategy that puts the focus squarely on customer relationship management.
“Today we have players whose day one proposition is services. The only thing these players own is their customer database and a channel to market, such as a call centre doing in-bound lead closure. Their aim is to identify and meet the needs of the most profitable strata of customers,” he says. In such a world, shouldering all the costs of owning a broad footprint, national or trans-continental network starts to look like an impossible debt burden. Incumbent telcos now have a battle on their hands preventing new entrants from cherry picking their most profitable customers and leaving them only with low margin or unprofitable customers.
“There needs to be a piece of the telco business that owns CRM and brand management. The role of companies like IBM is to help telcos understand their assets and to help them realign themselves so that they can market and sell the assets they have more efficiently. CRM is key to this,” Stevenson argues.
Brokering the bandwidth: how much do you want to pay?
Bandwidth brokering is now a highly active business in its own right. From companies laying reams of dark fibre on spec., to interactive web-based bandwidth trading exchanges like Band-X and others, the idea of providing organisations with the ability to shop around for the best price/performance deals is now well established in the sector.
Barry Flanigan, an Ovum analyst with responsibility for analysing bandwidth margins points out that on the popular European routes right now, such as the London to Frankfurt, or Amsterdam to Frankfurt routes, there is already far more fibre in the ground than can be justified by current levels of demand. “Today there are at least 20 major pan-European networks either in service, being built, or planned to roll out,” he says.
This means that companies at present have a railroad building mentality and are focusing on getting reach and announcing their presence. Within the next year they are going to have to start generating real revenue streams, and bandwidth brokering is going to help to ensure that prices keep going down. Network operators with unrealistic business targets will become ripe acquisition targets for larger players with deeper pockets.
In Flanigan’s view the most likely scenario is that the US RBOCs (regional Baby Bell operating companies), currently banned from the long distance market, will swarm into Europe, hungry for acquisitions, as soon as this ban is lifted. Together with incumbent operators they are the most likely to have the established revenue streams to bid on failed dark fibre roll outs. “We see the number of European fibre-owning companies slimming down to four to six players, with a similar scenario also playing out on a global scale,” he comments.
The new entrants: virtual everything and bags of service
The ability to lease bandwidth on existing networks and to either co-locate at the provider’s PoP (point of presence) or to build adjacent facilities, makes it very easy for new entrants to acquire a substantial, country specific reach very quickly.
However, as Genesis managing director Alan Barr observes, new entrants are under huge pressure to find a value proposition fast. His view, along with most people’s in the industry, is that this will come from next generation IP services, such as IP Virtual Private Networks, which give multi-site or multinational corporates a virtual unified data and voice network with more features and for a fraction of the cost of leased line networks. Part of the answer too, he suggests, will come from telco equipment manufacturers such as Genesis providing a good deal more by way of the ability for service providers to “groom” calls. “Banks, for example, need to be able to identify that a particular call is coming from a valued customer. Much of what is passed to the bank’s system, along with the call, could come from more intelligence on the network.
Bob Cushing, group marketing and business development director at Redstone Telecom, points out that his company, like many new entrants, has the SME market firmly in its sights. “We see added value as being all the services that SMEs could not afford to fund up-front, such as call centre technology. Delivering services on a pay as you go basis is going to be an extremely strong added value proposition for our market,” he notes.
Value added services: it comes with pictures
Aside from IP VPNs and ASP type services, one of the biggest value added services telcos and ISPs are moving towards is streaming content. From a telco’s point of view streaming video, multi-channel TV and MP3 audio services are wonderful bandwidth hogs that help to pump up the volume of traffic on their networks.
As Graham Seabrook, founder and CEO of the streaming media server company Ridgeway, explains, his company began life two years ago as a venture capital funded start-up, specialising in IP communications. “We decided that voice over IP would not be enough of a proposition, so we have focused on building streaming video server platforms for ISPs,” he says.
This involves allowing the ISP to provide both business and retail customers with a way of downloading client software, and using desktop cameras provided by the ISP, for video calling. “The reality today is the phone. Our approach enables an ISP to add video conferencing on top of the plain old POTs service. People can use the phones to talk and they can data share or look at each other on the PC,” he says. That way, one of the parties can be without a desktop camera, and they can still see the other on their PC or mobile device. This kind of new wave added value service, he reckons, is going to be one of the primary drivers of bandwidth demand going forward.
The impact of wireless content warehousing and Dick Tracy wrist watches
The advent of broadband wireless communications, expected to arrive in the UK later this year, will of itself be a huge stimulus to fixed network traffic demand. Matthew Finnie, group director of IP strategy at the pan-European network provider, Interroute, makes the point that one of the key factors boosting fixed wire demand will be the low cost of wireless devices. “Because you are looking at devices costing probably under £100, instead of the £600 plus figure for PCs, we are going to see access to broadband services becoming available to a much wider market. For us and for other carriers the result is going to be a huge increase in traffic on our networks as wireless operators look to download large chunks of content to feed to their customers,” he says.
Storm Telecommunications’ Wood says that one of the natural developments of the conjuncture between increased fixed wire bandwidth and broadband wireless, will be the development of local content warehouses. Traffic from these warehouses, which may contain hundreds of terabits of movies, will flow over fixed wire networks before being streamed over wireless connections to millions. Wireless, he notes, is excellent news for fixed wire network operators, both as a “last mile” access technology and as a driver for bandwidth consumption.
What the analysts say: the times are a’ changing
According to Ovum analyst Flanigan, we are going to see a huge rise in demand by corporates for leasing “wavelengths” (where companies take a whole 2.5Gb “lander” for themselves on a local, pan-European or trans-Atlantic route). “Wavelength services are a very hot area now. It is a form of virtual dark fibre leasing, where what you buy is a virtual fibre that has the capacity to be upgraded endlessly, as DWDM technology expands. Right now DWDM services are quadrupling in capacity every year, so this gives the larger corporates, who would normally be buying leased line global networks a very flexible strategy.”
Lars Gordell, European telecoms analyst with Forrester Research has spent the last three years researching telecoms markets in 17 countries, mostly in Europe. He points out that one of the problems – and challenges – telcos face is in finding attractive content that will give Europeans a reason to log on to their networks. “We found that over 70 percent of Europeans, including a significant proportion who already own PCs, are not interested in the Internet or on-line services because they do not feel that there is sufficient content there to hold their attention,” he says.
Gordell notes that a Nortel-chaired industry consortium, the Broadband Content Association, is looking at a whole range of content related issues. However, he points out, it might have made more sense if telcos and broadband network operators had put more thought into the content issue before they invested so much money in building out their networks. “We are looking at a very overcrowded market, with massive supply capability, facing a very thin demand level. The result is going to be huge pressure on pricing, which will spread round the globe pretty quickly,” he notes.
Gordell’s guestimate (he is in the middle of a report to be published in June) is that residential broadband penetration will not exceed 25 percent in the leading countries, and 10-15 percent of homes in the more backward markets of Europe, by 2004. Penetration of the SME sector across Europe, however, will be very much higher, he says.
Why not connect to…
Regularly updated telecoms news site
Quarterly review including overview, regional focus, industry issues and mobile internet
Telecoms link page