What gave me the most to think about was his emphasis on wireless. The topic cropped up several times in his talk, and Kriens clearly considers it a cornerstone of Juniper’s future. Take a look at Juniper’s Website, where “Mobile” gets equal billing with “Core” and “Edge” in the center of its home page.
I have to say that I agree with Kriens. Wireless is beginning to look as though it could be the hottest telecom trend in 2002 – and could end up driving developments in optical networks.
The importance of wireless hasn’t been obvious to me until fairly recently. This is partly because I’ve been preoccupied with optical technologies for the past 11 years. It’s also because developers of broadband wireless equipment have a record of big smoke and no fire.
All the same, things are changing. The deployment of IEEE 802.11 wireless LAN technologies to support broadband access services is exciting. So are developments that could see carriers finally making a significant move to voice over IP (VOIP), not for their wireline voice services, but for their wireless services. At the same time, the use of wireless data devices appears to be snowballing, and this is what optical networks need to hear: that new revenue-bearing sources of bandwidth are coming. Even GPRS phones, with their narrowband data limitations, represent a potentially huge source of bandwidth, just in the volume of users online at any one time.
According to IDC, there are about to be more wireless than wired devices in the world that can access the Internet. This bit of data seems portentous. It certainly suggests that as wireless devices increase in their usage, capacity demands, and application types, designers of Internet gear will more and more have to take into account the demands of wireless users.
From a capacity perspective, that effect is small today, but from the perspective of security and service management, it is quite large. But bandwidth will come along, too, perhaps not in the promise of 3G Wireless – which to date seems more like typical land-grabbing for spectrum than a guarantee of broadband services – but perhaps in yet another generation – “4G” wireless LAN technology adapted for the last few hundred feet.
Eventually service providers will find ways to make the mobile experience as near to wireline as possible. The enablers are coming together in the form of:
- The success of Blackberry Wireless devices, which despite having to type with one’s thumbs are incredibly appealing to millions of users and have become an essential appendage of the financial sector. The platform, and that of the PalmOS, is primed to exploit higher-speed services.
- The migration from circuit-switched cellular telephone networks to packet-switched, which underpins the 2.5G and 3G wireless networks. Sure, these may not be all they’ve promised, but that’s a matter of poor economics of building out these networks, not the fundamentals of the technology. The shift to packet-based cellular networks is key, whatever the fate of 2.5 and 3G wireless may be, because it transforms the wireless device from one that accesses the Internet to one that is perpetually connected to the Internet, an important distinction that greatly increases the variety of applications it can support.
- An improving base of infrastructure software for wireless data applications. This is obviously the hardest part of the equation, but it’s coming along well, with an increasing number of suppliers of the necessary application software, infrastructure middleware, gateway software, and enterprise platforms. In many ways, the Achilles heel of this enabler to watch will be security and privacy. The mobile Internet can’t flourish without them.
- The use of location-based technology to create a unique suite of applications ideally suited for mobile users. This allows the mobile device to not simply be stuck being a small-screened counterpart to a desktop system, but one that exploits its mobility to provide information to users necessary or useful to them at their specific locations.
- The Transformation of the Big Systems Integrators. The large vendors looking to reinvent themselves must do it in the context of a wireless world. Juniper formed an alliance with Ericsson AB (Nasdaq: ERICY); Nokia Corp. (NYSE: NOK)
bought Amber. Cisco Systems Inc. (Nasdaq: CSCO)
sells lots of IP phones now to the enterprise. Nortel Networks Corp. (NYSE/Toronto: NT)
and Lucent Technologies Inc. (NYSE: LU)? If they decide it's time to start spinning off businesses and shedding others, what they do with wireless is probably more important than what they do with optical – and it seems clear that no segment of equipment can be considered in isolation.
- The Next Generation of Routers. The design of routers must take into account the hundreds of millions of IP devices roaming the world of the near future. This will strain BGP (Border Gateway Protocol) to a bursting point, crowd the address space of IPv4, and push QOS up another degree as carriers begin carrying wireless voice over IP. The influx of bandwidth will further drive the need for terabit-level scaleability at the core, while increasing the types of service features required at the edge.
- The Transformation of Local and National Carriers. How the major carriers evolve their wireless networks will in large part determine who wins or loses in the coming decade of service provider competition, as wireless will represent a large and expandable revenue stream that is only beginning to be tapped. The lesson from the switched wireline voice network is that once a very stable telephone network was in place with a well standardized signaling system underpinning it, carriers could start adding layer upon layer of value-added services to basic voice, which has driven revenue increases and margins over the past decade. The same holds true for mobile, with an even greater promise for revenue through the integration of multimedia applications.
- The Last-Mile Bottleneck. As regulation reform moves at a snail’s pace in Congress, with carriers pitted against each other to control the last mile infrastructure, along comes a new service-provider-hardened wireless LAN technology that sits on a neighborhood telephone pole and gives the whole block high-speed Internet access. Any device with a $99 WLAN card attached can log on, speeding the arrival of home networks and, in some cases, community networks. Backhaul can be established with free space optics or broadband point-to-point wireless to make the whole network unbound. This could succeed where other attempts at broadband wireless access have failed, simply because this time it takes advantage of a low-cost enterprise technology, just as Ethernet access succeeds in the metro for similar reasons.
— Scott Clavenna, Director of Research, Light Reading