IPTV & Its Network Impact
After languishing for years as a quirky, niche technology, all of a sudden IPTV is generating some of the industry’s loudest buzz. Yes, the potential is huge. Although networks optimized for IP video services are being deployed today on a relatively small scale, primarily in Europe and Asia, the planned North American rollouts are poised to help fuel the expected tenfold subscriber growth over the next several years.
However, the IPTV market also presents some significant challenges for service providers. First of all, they need sufficient access bandwidth in place to support it. The RBOCs have made it clear that GPON is the technology of choice for many of the next-gen broadband access buildouts, and vendors such as Alcatel (NYSE: ALA; Paris: CGEP:PA) will look to evolve their ADSL technology with next-gen products in support of this trend. As service providers pursue the goal of boosting bandwidth past 20 Mbit/s per home, the impact on the aggregation and IP service edge network will be dramatic. The network, in short, will have to become “smarter.” There is still much debate about how to distribute intelligence throughout IPTV networks.
Most vendors have similar, but never identical, ideas about how to solve this problem. Dominant DSLAM vendors such as Alcatel want intelligence in the DSLAM. Core IP specialists such as Juniper Networks Inc. (Nasdaq: JNPR) and Cisco Systems Inc. (Nasdaq: CSCO) would rather keep more intelligence in the edge routers. The bottom line is, there are many different ways to design an IPTV network.
Some vendors are pushing Ethernet/IP DLSAMs, high-density Gigabit Ethernet aggregation switches, and video-centric broadband service routers/B-RASs. The horde of vendors going after these markets includes Alcatel, Cisco, ECI Telecom Ltd. (Nasdaq/NM: ECIL), Juniper, Redback Networks Inc. (Nasdaq: RBAK), and Riverstone Networks Inc. (OTC: RSTN.PK). Several themes will be consistent: higher system capacity, higher-density Gigabit Ethernet interface support, and high-availability architectures in support of new real-time service requirements. The network edge market is poised for a shakeout, and IPTV is the application that will cause the tremors.
— Rick Thompson, Senior Analyst, Heavy Reading
Ethernet Services & Equipment
If you're looking for a bright spot in the telecom industry, look no further than what's happening in Ethernet services. Carriers of all stripes have said in recent months that they have been experiencing double-digit year-over-year growth in Ethernet revenues. And in just the last few weeks, three carriers – one national operator and two regional players – told Heavy Reading that they are on track to blow past 2005 Ethernet revenue targets and have had to reset the sales bar higher for the rest of the year.
One of the unfortunate side-effects of all this growth is that it is fueling an explosion in the number of Ethernet service offerings, which will inevitably jack up competitive pressure. At Supercomm, you'll hear operators looking for equipment vendors to help them deal with these pressure by extending their service reach to more customer locations, delivering higher-performance services, supporting a wider range of value-added applications, and controlling network infrastructure and operating costs.
I expect quite a few carriers to highlight deployments or planned deployments of multipoint Ethernet services, given that this is the hottest part of the market right now. These services can be delivered using Ethernet switches/routers with MPLS and VPLS features or Sonet/SDH systems with RPR functionality. As for switches and routers, Cisco Systems Inc. (Nasdaq: CSCO) obviously plays strongly here, but the names to watch are really Alcatel and Tellabs Inc. (Nasdaq: TLAB; Frankfurt: BTLA), both of which have been gaining good traction with feature-rich switch/router products.
On the RPR front, while relatively few operators – including AT&T Corp. (NYSE: T), Cablevision Systems Corp.’s (NYSE: CVC) Lightpath, and PPL Telecom – are deploying the technology today, interest has picked up over the past six months from RBOCs and others. Cisco and Nortel Networks Ltd. (NYSE/Toronto: NT) are strong RPR players, while Lucent Technologies Inc. (NYSE: LU) and others are fighting for market share.
There has also been an explosion of interest in Ethernet access technologies in the past year, with dozens of operators evaluating platforms that support Ethernet over fiber and/or copper. Expect a lot more of this at the show. Carriers are interested in using these relatively cheap solutions to extend their Ethernet networks, control service operating costs, and provide a common and ubiquitous network. Startups seemingly generating the most buzz in this market include Anda Networks Inc., Covaro Networks Inc., Hatteras Networks Inc., and Overture Networks Inc. (which just announced a partnership with Juniper). Also making noise are Actelis Networks Inc., Ceterus Networks, Metrobility Optical Systems, MRV Communications Inc. (Nasdaq: MRVC), Paradyne Networks Inc. (Nasdaq: PDYN), RAD Data Communications Ltd., Verilink Corp. (Nasdaq: VRLK), and other creatively-named companies.
— Stan Hubbard, Senior Analyst, Heavy Reading
MPLS Migration at the Network Edge
MPLS technology and its various applications present some of the toughest questions to be raised at Supercomm. The biggest question may be its use in access networks and just how critical this is to operators. More importantly, if they have a strategy and application in mind for the technology, how do they implement it? Heavy Reading has talked with more than a dozen service providers about this topic in the past three months, and we have found this to be one of the most daunting analytical tasks ever.
Operators are of many different minds. Some want to move customers off legacy services onto Layer 3 IP VPNs. Others want to preserve legacy services, but improve their margins by tunneling them through the IP/MPLS core using pseudowires. A third strategy is to preserve access circuits but evolve the service using a Layer 3 MPLS VPN structure in the core. Yes, the mind spins trying to form a useful opinion when advising vendors. Do everything! There is a session on pseudowire solutions Monday afternoon that could hold some answers, and we'll be seeing some of the latest multiservice switch and edge router solutions first-hand on the floor.
ROADM is the buzzword of the day in optical, but is anyone really deploying the stuff? The answer is certainly yes, but the hype created by large RFPs from SBC Communications Inc. (NYSE: SBC) and Verizon Communications Inc. (NYSE: VZ) has stirred the optical pot to the point of excess. Too many vendors are still competing for this rather small piece of the pie, which continues to suffer the enervating mix of unreasonable pricing pressure and advanced feature requirements - multi-degree wavelength switched nodes, with GMPLS, mesh networking, integrated Sonet switching, automated amplifiers, dispersion compensation, wavelength switching, performance monitoring, you name it, all at a price premium of zilch, zip, nada. Sounds like fun, eh?
Fujitsu Ltd. topped our recent ranking of ROADM system vendors, so Heavy Reading will be catching up with it, plus the other contenders, including Alcatel (and partner Tropic Networks Inc.), Lucent Technologies Inc. (NYSE: LU) (and partner Movaz Networks Inc.), Mahi Networks Inc., Meriton Networks Inc., and of course Cisco, which is quietly chipping away at the metro DWDM market with its 15454 MSTP platform. It'll also be worth seeing what ZTE Corp. (Shenzhen: 000063; Hong Kong: 0763) and Huawei Technologies Co. Ltd. have to offer. They are building momentum in the optical market and will undoubtedly have to be reckoned with.
There are a number of questions that will (hopefully) be answered at Supercomm this year. Such as: Now that RBOCs have all the regulatory relief they've ever asked for, what's next? Will their keynotes now call for the end of video franchise requirements?
How well is Tellabs Inc. (Nasdaq: TLAB; Frankfurt: BTLA) doing? It won the BellSouth Corp. (NYSE: BLS) business for FTTC; it continues to forge ahead at Verizon, winning business all over the world with its 8800 routers from the Vivace acquisition; and it continues to succeed with the 5500 DCS, which everyone counted out in 2000. Those wins seem pretty big, so why does it seem as though everyone – most noticeably Wall Street – has a sinking feeling about the company? Kind of a puzzle, and I plan to investigate first-hand. [Ed. note: Light Reading plans to investigate as well, including conducting an interview with Tellabs CEO Krish Prabhu.] Back at Supercomm 1993, the buzz was the presence of cable TV equipment suppliers, the promise of a telco "video dialtone," and the first wave of real convergence between telco and cable. This year, look for a new twist: No real mergers between telco and cable companies, but you'll still see Scientific-Atlanta Inc. (NYSE: SFA), Motorola Inc. (NYSE: MOT), and Harmonic Inc. (Nasdaq: HLIT) touting video gear for telcos. This year, the saviors of the RBOCs are IPTV and FTTP, but I can't help harkening back to 1993, when Bell Atlantic announced its intent to merge with TCI, then called it off six months later, then took a full decade to develop a new video strategy. Will this time around be any different?
Another big question: Are Nortel and Lucent really turning into wireless equipment suppliers? Well, since they are very involved in the VOIP infrastructure market, actually the answer is no; but it's still a question worth keeping in the back of one's mind as you sift through the pile of press releases from each vendor at this year’s Supercomm.
— Scott Clavenna, Chief Analyst, Heavy Reading
IP Multimedia Subsystem (IMS)
IMS has come out of nowhere in the past 12 months to emerge as one of the core technologies aiding the mobile network to migrate to IP next-generation networks (NGNs), and a whole parcel of IMS product announcements and enhancements, positioning statements and partnering deals will be unwrapped in Chicago (see IMS Guide).
Although IMS began life as a 3rd Generation Partnership Project (3GPP) cellular wireless technology (UMTS), it’s now been adopted by 3GPP2 (CDMA 2000); Cable Television Laboratories Inc. (CableLabs) (PacketCable); and the European Telecommunications Standards Institute (ETSI)’s TISPAN (wireline telco networks). At Supercomm, the Alliance for Telecommunications Industry Solutions (ATIS) will be setting out its own plans to incorporate IMS. It looks set to be the core technology for integrating SIP-based applications across both wireline and wireless networks. The prospect of using it to lower development costs of a wide portfolio of applications has really captured the imagination of the big telco community.
Some leading IMS software and hardware vendors have taken an early lead and will be looking to consolidate their positions at Supercomm, while everyone works hard to catch up. The leaders include Ericsson AB (Nasdaq: ERICY), Lucent Technologies Inc. (NYSE: LU), and Siemens AG (NYSE: SI; Frankfurt: SIE). But establishment heavyweights Alcatel (NYSE: ALA; Paris: CGEP:PA) and Nortel Networks Ltd. (NYSE/Toronto: NT) have also set out ambitious roadmaps for IMS; while Huawei Technologies Co. Ltd., Motorola Inc. (NYSE: MOT), and Nokia Corp. (NYSE: NOK) each hopes to use IMS to widen its market reach.
Many service providers believe the biggest attraction of IMS will be its use of standard open interfaces that allow them to pick best-of-breed components at every level. Among other things, that’s allowing a raft of IT specialists a much better route into the telecom applications and network systems integration markets. IT vendors following this path include BEA Systems Inc. (Nasdaq: BEAS), Hewlett-Packard Co. (NYSE: HPQ), IBM Corp. (NYSE: IBM), Intel Corp. (Nasdaq: INTC), Microsoft Corp. (Nasdaq: MSFT), and Sun Microsystems Inc. (Nasdaq: SUNW). We expect all of these companies will be trumpeting their IMS credentials in Chicago.
But IMS also provides openings for many smaller players, some of which have already made a big splash. They include applications server specialists like Ubiquity Software Corp., and media server vendors like Convedia Corp. Finally, Chicago will see softswitch vendors like Sonus Networks Inc. (Nasdaq: SONS) and Veraz Networks Inc. setting out transition strategies for IMS.
For more see: IP Multimedia Subsystems: Easy Does It and IMS Guide.
Mobile Next-Gen Networks
Distributed switching is the most “here and now” market in the mobile core. It mirrors the move to softswitch architectures in the wireline world, but adds a twist of mobility. In mature markets the issue is an ageing installed-base of mobile-service switching centers (MSCs), which are expensive to operate and maintain in the context of rapidly increasing traffic but slowing subscriber and revenue growth. In emerging markets the driver is, quite simply, exploding subscriber growth. In both cases the requirement is the same: inexpensive, scaleable, voice switching over an IP core.
From the vendor side of the house, it’s imperative to have a competitive play here – we’re talking about the next 10 years of mobile voice. And let’s face it: With more and more people abandoning wireline telephony, more and more will this become the serious end of the telecom business. Recent notable events in this segment include Alcatel’s $250 million acquisition of Spatial Wireless last year, and Huawei’s emergence on the scene. Nokia, Ericsson, Motorola, et. al., are also in the game and selling to win. The secondary pitch is that mobile NGN is an essential step towards a fully-loaded IMS architecture.
For more on this, see the upcoming Unstrung Insider report: “The Future of Mobile Voice: Softswitch MSCs.”
Mobile Packet Core
The major trend in the packet core is the development of “service-aware” Layers 4-7 packet processing. The motivation is that mobile operators want, at all costs, to avoid becoming dumb pipes to the subscriber. A parallel development is the evolution the GGSN (a.k.a. “mobility router”), which is being beefed up to handle greater traffic loads and to host the “service aware” processing functions. In some cases the GGSN and SGSN are being combined in one unit (xGSN).
Examples of products to watch here are Nokia’s Integrated Services Node, Alcatel’s iGGSN, and NEC Corp.’s (Nasdaq: NIPNY; Tokyo: 6701) xGSN. At some point, expect fruit from the Siemens/Cisco partnership and look for more service-aware capability from the Ericsson/Juniper Networks Inc. (Nasdaq: JNPR) combo. In the CDMA world, Starent Networks Corp. is blowing the doors off and winning in just about every major account.
There are two other fundamental reasons these boxes need more upgrades. The first is higher-speed radio access networks – which, if you believe vendor claims, could soon deliver 1-Mbit/s, low-latency connections to the subscriber. The second is that these nodes could become the convergence “on-ramp” to the services domain for wireless LAN – and maybe even broadband – users.
For more, see: NEC Scores With ATCA Core and Nokia Unveils Convergence Kit.
Wireless LAN and Convergence
Will convergence fix your kitchen sink? What does it mean anyway? In any case, it’s sure to be in the air at Supercomm. The most important element of convergence, from a wireless perspective, is how to bring wireless LAN access into the mobility domain. More broadly it could be defined as converging mobility with any IP access network.
Convergence is, of course, easier said than done, and, for the most part, the cellular world doesn’t get it… yet. This reticence is understandable. After all, mobile operators hold most of the trump cards today, and the issues around developing and shipping dualmode WLAN/cellular phones are undoubtedly immense – especially if you figure on making a profit. But longer term, the explosion of wireless LAN and VOIP at home, in businesses, and in the community, is going to force their hand. NTT DoCoMo Inc. (NYSE: DCM) is showing the way in Japan, while T-Mobile USA is said to be trying to figure out a way to introduce unlicensed mobile access technology in the States. Plenty of vendors, mostly startups, are lining up to help. For more, see: The Third Way of Convergence and Mobile's Core Competence.
Advanced Telecom Computing Architecture (ATCA)
Expect to hear a lot about ATCA at Supercomm – again! The reason is simple. ATCA vendors need to maintain the momentum they’ve managed to generate behind this emerging standard if they’re ever going to make a profit on it. And OEMs need to publicly, and loudly, pledge their commitment to ATCA to encourage the supplier base to stick with the program and continue investing. More than anything, OEMs need a competitive supplier base if they’re to reap cost-savings from ATCA.
Beyond the show floor, ATCA does have good traction among vendors of mobile core network equipment., Alcatel, Huawei, Motorola, NEC, Nortel, and Siemens are all committed to port various applications to the new platform, and it wouldn’t be a surprise to see Lucent follow suit. Even vendors that see ATCA as being yet a ways down the road – Nokia and Ericsson, for example – are committed to blade server architectures for the mobile core and IMS service layer, and will possibly adopt complementary PICMG standards, such as advanced mezzanine cards. The logic behind modular designs, using off-the-shelf components, is compelling.
For more on ATCA, see: ATCA: 3G Cost Cutter and The ATCA Chain Reaction.
— Graham Finnie, Senior Analyst, Heavy Reading, and Gabriel Brown, Chief Analyst, Unstrung Insider