Generation W

The market for consumer-oriented Wi-Fi WLAN products is exploding. Starting from nowhere in 2000, Linksys captured more than 11 percent of the global wireless LAN market in 2002, catching the eye of Cisco Systems' acquisition brain trust in the process.

The Wi-Fi hotspot market also is burgeoning, as a complex value chain of infrastructure and service providers promises to deliver wireless broadband data services to a location near you.

But despite all this kinetic motion, the enterprise WLAN market has remained nearly flat. A modest increase in unit volume in 2002 was offset by a decline in dollar value, to about $800 million, according to Synergy.

Why? Faced with demand emanating from the executive suite as well as business units, IT professionals are struggling--not so much with whether they should deploy, but with what to deploy and when. Underlying standards issues create uncertainty regarding investment protection, security problems are legendary, and manufacturers have failed to provide true enterprise-class WLAN systems. But that's all about to change.

In 2004, we'll see new products, together with--fingers crossed--the resolution of some significant standards issues (see "WLAN Standards Watch,"). For the first time, an IT manager will be able to deploy a scalable, secure, manageable WLAN that can serve as a high-availability infrastructure for both conventional applications, such as Web access and e-mail, and emerging functions, including wireless voice over IP.First-generation WLANs were proprietary and targeted at high-margin vertical applications, like retail and warehousing. Many were viewed as operational support systems outside the province of IT, and few network engineers bookmarked the Web sites of leading market players like Aironet Wireless Communications, Proxim and Symbol Technologies.When the IEEE 802.11 standard was approved in 1999 and the Wi-Fi Alliance was established to promote 802.11b through marketing and product certification, the industry began its second-generation progression toward mainstream product sets. The so-called "smart" access point was a the key element of WLAN infrastructure. These APs provided both physical-layer radio modems and Layer 2 network services.

For the past five years, APs have grown increasingly powerful. But deploying and managing networks built around hundreds or thousands of APs is costly.

Moreover, the smart AP architecture has proved inadequate, giving rise to various third-party enhancements. AirWave Wireless and Wavelink Corp. developed software-based management systems that beat those provided by AP manufacturers. Bluesocket, Vernier Networks and ReefEdge stepped in to address fundamental security, roaming and management issues. AirDefense, AirMagnet and Wild Packets introduced monitoring systems. The list goes on.

Now, third-generation enterprise WLAN systems are emerging that provide a more comprehensive integrated solution--not only from established vendors like Cisco, Proxim and Symbol Technologies, but from new players--Airespace, AirFlow Networks, Aruba Networks, Chantry Networks, Legra Systems, Trapeze Networks and Vivato--as well as established wired infrastructure players like Extreme Networks and Nortel Networks.

These 3G enterprise WLAN products, often marketed as "wireless switches" to lure comparisons to the extremely valuable migration of Ethernet from shared to switched, provide a similar range of services and, in many cases, underlying system architecture.Naturally, Cisco's role in the market is key. If Cisco succeeds in maintaining--or even increasing--its dominant enterprise WLAN market share, many of the companies listed above will disappear. However, Cisco is clearly vulnerable, for two reasons. First, its existing architecture has limitations. The company understands this and is in the midst of a complex evolution towards SWAN, short for Structured Wireless Aware Network. This new architecture leverages a primary Cisco asset, with APs running IOS and more fully integrated with Catalyst switches, which will be equipped with wireless extensions to their native suites of network services. To top it off, there is a management overlay, known as the CiscoWorks Wireless LAN Solution Engine, or WLSE. If successful, Cisco will deliver an integrated set of WLAN services. But this brings us to the second bump in the road: Getting to this integrated nirvana will require careful coordination among multiple Cisco business units, something for which the company is not noted.

The venture capital community doesn't think Cisco will succeed in establishing greater dominance, evidenced by the level of funding flowing to wireless start-ups. VCs aren't just betting that enterprise IT managers will deploy WLANs in large numbers, they're wagering that IT managers will recognize the value proposition offered by the start-ups being funded. VCs believe IT professionals--even those with Cisco wired networks--are willing to consider new technology partners focused on delivering a better wireless mousetrap.

The key word here is better. Some early attempts at WLAN design innovation weren't well-received. Proxim's Harmony system, for example, provided a more integrated approach to managing a WLAN infrastructure, but it lacked scalability and advanced radio management, mobility, security and system monitoring capabilities.

Symbol followed with Mobius (now known as Symbol Wireless Switch System), which used inexpensive thin APs to off-load Layer 2 processing to a closet switch. Symbol's market message was compelling: By moving intelligence from the edge to the core, the company promised to deliver a low-cost infrastructure with a comprehensive array of easily managed wireless services. This was a risky move for Symbol; but though it has taken significantly longer than the company projected to deliver a stable system, Symbol estimates that 40 percent of new system sales are for the new architecture.

Symbol has run up against significant competition from a number of start-ups, most of which are well-funded and staffed by seasoned industry veterans. Most give Symbol credit for a good idea, but claim that their own systems are far more capable.Because there are so many start-ups in this market, vendor viability is a huge issue. And, not surprising, it's a sensitive topic. There's not as much talk about these newbies being acquired by larger competitors. There just doesn't appear to be a large number of potential suitors.

What makes this market interesting is the sense that one or several of these start-ups could become leaders in a rapidly expanding enterprise WLAN market. That's the promise that keeps all those network engineers at work on weekends.

As you'll learn from our review of WLAN infrastructure products (see "New WLAN on Campus,"), we explored the capabilities of these new wireless infrastructures in depth, comparing them with market leader Cisco and came away convinced of two things. One, these systems represent a dramatic improvement over earlier offerings, providing compelling benefits for enterprise IT and an interesting market opportunity for new companies.

Two, which may comfort those of you who prefer a conservative approach, is that getting it right won't be all that much easier for the start-ups, which face enormous technical, marketing and channel issues, not to mention the prospect of facing off with the industry leader for the favor of enterprise IT management. Don't expect to hear many IT managers talk about replacing their Cisco infrastructure with something new.

A recurring promise from almost every WLAN vendor is: "Easy and seamless integration with your existing network and systems environment." Not even the most brazen marketer can expect experienced IT professionals to take that pledge seriously. These are complex systems and version 1.0 offerings. This stuff is brand-new.Many of us like the smell of a new car. Few like the smell of new software.

But don't conclude it's all gloom and doom for the new vendors. Based on the number of high-profile customer engagements touted by the start-ups, many enterprises appear willing to test the waters with tactical pilot deployments of new products from unproven companies. They've waited this long to make a more strategic decision, so why not wait a little longer?

That makes 2004 a critical year for the WLAN industry, even if it's not the year during which all our wireless problems will be solved. As start-ups gain field experience, the inevitable bugs will be eradicated. Vendors will battle not only to provide a core suite of basic services but to differentiate themselves from the pack with unique features. As for Cisco and other incumbent providers, they'll have their chance to prove that evolution makes more sense than revolution when it comes to the critical needs of modern enterprises.

Dave Molta is a senior technology editor at Network Computing. He is also assistant dean for technology at the School of Information Studies at Syracuse University and director of the Center for Emerging Network Technologies. Molta's experience includes 15 years in IT and network management. Write to him at [email protected].

Post a comment or question on this story.

Have fears of your WLAN gear turning into a legacy albatross kept your mobile implementations localized--or worse yet, nonexistent? We sympathize. Whenever a technology's standards are in flux, as are those for 802.11 (insert letter du jour), it's tempting to stay on the sidelines.But now's the time to get serious: 2004 will usher in a more mature generation of products, and we expect some major standards to be nailed down. Prices will continue to drop, and innovation will be the name of the game. Plus benefits will go beyond the usual mobile Web and e-mail access. How does wireless voice and video over IP sound?

Serendipitously, just as we were gearing up for the mother of all WLAN reviews, we found out that Syracuse University's School of Information Studies was renovating its four-story, 50,000-square-foot Hinds Hall digs and planning a full wireless overlay with 802.11b and 802.11a support. So we worked up an RFP using this scenario, and asked vendors to bring their gear on site for hands-on testing. We stressed security, including flexible authentication, access control and rogue-device detection; reliability; scalability; ease of management; and investment protection. And to spice things up, we threw in wireless VoIP, as well as video and transparent mobility across IP subnets.

A mix of established providers--Cisco Systems, Enterasys Networks and Symbol Technologies -- and start-ups -- Airespace, Aruba Wireless Networks, Chantry Networks and Trapeze Networks--committed the resources to participate. After exhaustive testing, we gave the well-rounded, appliance-based Airespace Wireless Enterprise Platform our Editor's Choice award.

WLAN standards are in a constant--albeit glacially slow--state of change. In some respects, this painfully deliberative pace is a result of the complexity associated with the underlying WLAN issues. It's not easy to design security and QoS (Quality of Service) standards for wireless environments. It's also true that there's a rich history of vendor politics associated with standards, with big industry players often willing to obstruct when directions don't serve their corporate interests.

Despite these problems, some progress can be noted. For example, the 802.11b, 802.11a, and, to a lesser extent, 802.11g standards are solid, with high levels of interoperability across products that use different radio chipsets. However, other important standards are still being hammered out. Take a look at the chart for a rundown.More than $1 billion in venture capital has been invested in the Wi-Fi market since January 2001, according to a recent report by Fierce Wireless and the Wireless Data Research Group. Clearly, a significant number of savvy investors think wireless is the next big thing.Much of the early backing went to wireless chipset developers, a market that is now overcrowded and in the early stages of consolidation. More recent VC investments have targeted vendors of wireless systems and services, including several of those represented in our RFP analysis.

Some related findings:

• About one-third of this funding went to wireless chip companies.

• Another third went to companies focused on network operations, including software and hardware infrastructure products.

• More than $100 million was invested in the hotspot market, with two-thirds of that going to service providers/network operators and the remainder going to software and hardware companies developing products for this market.• Only about 6 percent of VC funding went to companies building wireless applications, most focused on voice-over-wireless LAN systems.

• Intel Capital was the most prominent VC company, making at least 15 private investments during the 18-month period addressed by the Fierce Wireless study.