When cable operators get their hands on DOCSIS 3.0-compliant equipment, which could happen within a year if you are to believe some, telcos suddenly will be faced with a competitor that can offer high-speed access services faster than anything they've ever faced. By combining, or “bonding,” channels within the spectrum available on coax, cable operators, in fact, will be able to offer Internet access services up to 100 Mb/s.
And while virtually everyone agrees that telcos will need to respond to such a looming threat, there is widespread disagreement on how to do that. Some vendors, pointing to the experience in Japan and Korea, where 100 Mb/s is available via VDSL, say telcos need to match cable operators in this speed race. Others believe the solution lies in marketing and bundling services.
Perhaps the most immediate question telcos must answer is whether they can match a 100 Mb/s service from cable. The answer might not be readily obvious. With a fiber-to-the-home (FTTH) architecture, the technical capability is available. However, most telcos with FTTH deployments aren't offering 100 Mb/s data service because of the cost of equipment and transport at the aggregation points. Verizon, for instance, is limiting its data service to 15 Mb/s on its FiOS network.
The standardization and future deployment of VDSL2 also will give carriers the ability to offer data service in excess of 50 Mb/s, though most are expected to offer something closer to 10 Mb/s.
Because while the technology can speed data down fiber and copper, IP-based video going down the same path has an even bigger impact on the equation. As telcos move into providing video service, the bandwidth requirements jump significantly. Using the most advanced available compress rates (and staying in the good graces of content providers that are justifiably concerned about picture quality) each standard definition stream takes at least 2 Mb/s out of the bandwidth budget, and each high-definition streaming requires a bare minimum of 9 Mb/s. That in and of itself begs the question of how many video streams will be provided and whether any of them will be high-definition streams.
In Asia, where some carriers are able to offer a triple play that includes video and very high-speed Internet access, one video stream typically is sufficient. However, in North America, most vendors say they rarely see a request for proposals with anything less than three video streams.
“The minimum is three streams per household in North America, and you usually could have one of those streams being high-definition,” said Fabrice Beer-Gabel, senior director of product marketing for telecom video for BigBand Networks. “That line also has to carry data, and we're seeing some real inflation on those data rates. We only have to expect that broadband services won't limit themselves to 1 Mb/s. Fairly quickly, you get to the point where 15 Mb/s gets pretty tight.”
Indeed, most telcos have plans in place that will put their total bandwidth budgets for the triple play well beyond 15 Mb/s, though most don't anticipate going past 50 Mb/s.
During the development of the VDSL2 standard, both BellSouth and SBC Communications were vocal about wanting a technology that would allow them to offer a total bandwidth of around 30 Mb/s. Although not getting close to the 100 Mb/s that cable operators will be able to offer, there is a significant segment of the market that believes telcos shouldn't get into a race for speed but should be heavily marketing the total package of services — and pointing out to customers that 100 Mb/s from a cable operator is still on a shared pipe.
“From the telco side, there are a lot of advantages,” said Ken Madison, senior product marketing manager of chip vendor Centillium. “If you're just talking about the bandwidth battles, those are going to be fairly well fought by telcos because of the introduction of VDSL2. One of the advantages of DSL is it's a point-to-point connection, so essentially you're getting the full pipe. That's a clear advantage that cable only can overcome by really pumping up the pipe to 500 Mb/s.”
At the same time, telcos also must step up their bandwidth to at least be comparable with cable, he added.
“Somewhere between 30 and 50 Mb/s of bandwidth is what's going to be used in this triple play,” Madison said. “SBC wants 30, but my personal feeling is that 30 Mb/s isn't going to be enough. I think it's going to settle closer to 50 Mb/s.”
Madison also points to the Japanese market as a potential harbinger of the coming high-speed battle. In the battle between NTT and Yahoo Broadband, the latter was offering a 47 Mb/s service for the equivalent of a few more dollars per month compared to NTT's 45 Mb/s service.
“You saw people signing up for the faster service even though they had no clue what they were going do with an extra 2 Meg,” he said. “After the dust settles, I think people are going to really start looking at what they are getting, though.”
Perception also can play a big part in telcos' marketing plans, said Ken Couch, director of marketing for broadband networks at Nortel Networks, which is advising carriers not to get into a battle of speed with cable operators.
“In the end, over the long term, that has little value to the consumer,” he said. “Do I care if my presentation uploads in 1 millisecond versus .001 milliseconds? The marginal benefits you get aren't that important.”
Instead, telcos carriers should put their efforts into the converged services model that integrates IPTV and other data elements. Not surprising, of course, is the company's pushing of the IP multimedia subsystem architecture, where services are interchangeable regardless of their delivery system. Cable companies simply can't match that flexibility.
Couch sites the example of carriers offering a sports package of services to subscribers. In that bundle there would be plenty of traditional cable sports networks but also the ability to link into customized information when away from home, say in a hotel with a broadband connection or in a mobile environment via wireless handset.
“That kind of service level to a consumer would be very attractive,” he said.
Likewise, as carriers move into that environment, consumers' concern over bandwidth will disappear. In the IPTV environment, for instance, users won't care how much bandwidth is being fed to their homes because it will be more than just video services that they are using, said Rich Cardone, director of marketing at Tazz Networks.
“There's a lot of stuff going on that is data related, but you might not even realize it because the interfaces are changing,” he said. “Buying a shirt you see someone wearing on TV, for example. What it comes down to is what you offering.”
Perhaps the other reason telcos may not want to get into a speed contest with cable operators is the cost. Because cable is inherently a shared medium, it doesn't typically need the same kind of bandwidth at the aggregation points. For telcos, DSL is a dedicated pipe and must be matched with similarly large bandwidth pipes at aggregation. In some cases, that adds a significant cost element, but the migration to Gigabit Ethernet uplinks from DSLAMs helps, said Don McCullough, director of product marketing for Entrisphere.
“Once you're on Ethernet end-to-end, you're on a different price curve,” he said. “What we've seen so far is we've been able to keep on the Moore's law cost curve about as fast as the telephone companies are beating another chunk of bandwidth out of us at the same cost.”
Even with the reduced cost, the emergence of on-demand video services is certain to push the bandwidth requirements up even further. Layer on top of that the possibility of Internet access in excess of 50 Mb/s, and you end up with a bandwidth demand that is almost exponentially higher than current requirements, said Marco Wonders, chief marketing officer of Redback Networks.
“There are technologies to ensure that the bandwidth at the edge is still limited. In IPTV, it's multicasting. The trend now, though, is that multicast is going to be big in terms of numbers of users. You will definitely start needing more bandwidth in the aggregation layer.”
http://telephonyonline.com/fttp/news/telecom_telcos_avoid_speed/
And while virtually everyone agrees that telcos will need to respond to such a looming threat, there is widespread disagreement on how to do that. Some vendors, pointing to the experience in Japan and Korea, where 100 Mb/s is available via VDSL, say telcos need to match cable operators in this speed race. Others believe the solution lies in marketing and bundling services.
Perhaps the most immediate question telcos must answer is whether they can match a 100 Mb/s service from cable. The answer might not be readily obvious. With a fiber-to-the-home (FTTH) architecture, the technical capability is available. However, most telcos with FTTH deployments aren't offering 100 Mb/s data service because of the cost of equipment and transport at the aggregation points. Verizon, for instance, is limiting its data service to 15 Mb/s on its FiOS network.
The standardization and future deployment of VDSL2 also will give carriers the ability to offer data service in excess of 50 Mb/s, though most are expected to offer something closer to 10 Mb/s.
Because while the technology can speed data down fiber and copper, IP-based video going down the same path has an even bigger impact on the equation. As telcos move into providing video service, the bandwidth requirements jump significantly. Using the most advanced available compress rates (and staying in the good graces of content providers that are justifiably concerned about picture quality) each standard definition stream takes at least 2 Mb/s out of the bandwidth budget, and each high-definition streaming requires a bare minimum of 9 Mb/s. That in and of itself begs the question of how many video streams will be provided and whether any of them will be high-definition streams.
In Asia, where some carriers are able to offer a triple play that includes video and very high-speed Internet access, one video stream typically is sufficient. However, in North America, most vendors say they rarely see a request for proposals with anything less than three video streams.
“The minimum is three streams per household in North America, and you usually could have one of those streams being high-definition,” said Fabrice Beer-Gabel, senior director of product marketing for telecom video for BigBand Networks. “That line also has to carry data, and we're seeing some real inflation on those data rates. We only have to expect that broadband services won't limit themselves to 1 Mb/s. Fairly quickly, you get to the point where 15 Mb/s gets pretty tight.”
Indeed, most telcos have plans in place that will put their total bandwidth budgets for the triple play well beyond 15 Mb/s, though most don't anticipate going past 50 Mb/s.
During the development of the VDSL2 standard, both BellSouth and SBC Communications were vocal about wanting a technology that would allow them to offer a total bandwidth of around 30 Mb/s. Although not getting close to the 100 Mb/s that cable operators will be able to offer, there is a significant segment of the market that believes telcos shouldn't get into a race for speed but should be heavily marketing the total package of services — and pointing out to customers that 100 Mb/s from a cable operator is still on a shared pipe.
“From the telco side, there are a lot of advantages,” said Ken Madison, senior product marketing manager of chip vendor Centillium. “If you're just talking about the bandwidth battles, those are going to be fairly well fought by telcos because of the introduction of VDSL2. One of the advantages of DSL is it's a point-to-point connection, so essentially you're getting the full pipe. That's a clear advantage that cable only can overcome by really pumping up the pipe to 500 Mb/s.”
At the same time, telcos also must step up their bandwidth to at least be comparable with cable, he added.
“Somewhere between 30 and 50 Mb/s of bandwidth is what's going to be used in this triple play,” Madison said. “SBC wants 30, but my personal feeling is that 30 Mb/s isn't going to be enough. I think it's going to settle closer to 50 Mb/s.”
Madison also points to the Japanese market as a potential harbinger of the coming high-speed battle. In the battle between NTT and Yahoo Broadband, the latter was offering a 47 Mb/s service for the equivalent of a few more dollars per month compared to NTT's 45 Mb/s service.
“You saw people signing up for the faster service even though they had no clue what they were going do with an extra 2 Meg,” he said. “After the dust settles, I think people are going to really start looking at what they are getting, though.”
Perception also can play a big part in telcos' marketing plans, said Ken Couch, director of marketing for broadband networks at Nortel Networks, which is advising carriers not to get into a battle of speed with cable operators.
“In the end, over the long term, that has little value to the consumer,” he said. “Do I care if my presentation uploads in 1 millisecond versus .001 milliseconds? The marginal benefits you get aren't that important.”
Instead, telcos carriers should put their efforts into the converged services model that integrates IPTV and other data elements. Not surprising, of course, is the company's pushing of the IP multimedia subsystem architecture, where services are interchangeable regardless of their delivery system. Cable companies simply can't match that flexibility.
Couch sites the example of carriers offering a sports package of services to subscribers. In that bundle there would be plenty of traditional cable sports networks but also the ability to link into customized information when away from home, say in a hotel with a broadband connection or in a mobile environment via wireless handset.
“That kind of service level to a consumer would be very attractive,” he said.
Likewise, as carriers move into that environment, consumers' concern over bandwidth will disappear. In the IPTV environment, for instance, users won't care how much bandwidth is being fed to their homes because it will be more than just video services that they are using, said Rich Cardone, director of marketing at Tazz Networks.
“There's a lot of stuff going on that is data related, but you might not even realize it because the interfaces are changing,” he said. “Buying a shirt you see someone wearing on TV, for example. What it comes down to is what you offering.”
Perhaps the other reason telcos may not want to get into a speed contest with cable operators is the cost. Because cable is inherently a shared medium, it doesn't typically need the same kind of bandwidth at the aggregation points. For telcos, DSL is a dedicated pipe and must be matched with similarly large bandwidth pipes at aggregation. In some cases, that adds a significant cost element, but the migration to Gigabit Ethernet uplinks from DSLAMs helps, said Don McCullough, director of product marketing for Entrisphere.
“Once you're on Ethernet end-to-end, you're on a different price curve,” he said. “What we've seen so far is we've been able to keep on the Moore's law cost curve about as fast as the telephone companies are beating another chunk of bandwidth out of us at the same cost.”
Even with the reduced cost, the emergence of on-demand video services is certain to push the bandwidth requirements up even further. Layer on top of that the possibility of Internet access in excess of 50 Mb/s, and you end up with a bandwidth demand that is almost exponentially higher than current requirements, said Marco Wonders, chief marketing officer of Redback Networks.
“There are technologies to ensure that the bandwidth at the edge is still limited. In IPTV, it's multicasting. The trend now, though, is that multicast is going to be big in terms of numbers of users. You will definitely start needing more bandwidth in the aggregation layer.”
http://telephonyonline.com/fttp/news/telecom_telcos_avoid_speed/
