Eighteenth century English poet Alexander Pope got it right when he penned "Fools rush in where angels fear to tread."
Then again, Pope never had the chance to meet an I.T. administrator faced with aging networks, high-bandwidth mobile applications and pressures to deploy more cost-effective technologies without slamming the operation of the current system.
Many I.T. and network executives tend toward the less-than-poetic approach in adopting seemingly revolutionary technologies: Waiting for swashbuckling early adopters to take the plunge before easing themselves into the water. Only a handful dive right in, sometimes with little regard for risk or return on investment.
The "better safe than sorry" approach usually works best, especially when newer technologies have the potential to adversely impact a perfectly good architecture, or may still be a few versions shy of a finished product. Don't think so? Just ask the people who decided to be pioneers when the first wave of pre-802.11n Wi-Fi products hit the market a year or two ago. Or, ask those people who thought early Bluetooth-enabled laptops were a safe bet so long as they were provided by a top-tier company.
The problem with waiting too long to at least dip your toes into a new technology is that you might put yourself at a competitive disadvantage when it comes to upgrading and expanding existing systems. This is especially true of 802.11n Wi-Fi, which promises considerable improvements in network speed, reliability and coverage over current wireless LAN alternatives.
"Most networks are five or more years old and are due for an upgrade," notes Rachna Ahlawat, VP of strategic marketing for Meru Networks and an ex-Gartner analyst who closely followed the networking space. Ahlawat says the quality-of-service (QoS) benefits and performance boost offered by the latest 802.11n products are a "whole different ballgame" as compared with 802.11 a/b/g technology, although she admits that adoption will not happen overnight.
WI-FI SPECS NOT FINAL
One of the major reasons why 802.11n Wi-Fi adoption will take time is that the final ratification of the specification won't be released to vendors until early 2009. Meanwhile, earlier this year, 802.11n Draft 2.0 was released. Products incorporating this near-final standard are being tested and certified by the Wi-Fi Alliance, a non-profit group that is charged with driving a single worldwide standard for high-speed wireless (www.wi-fi.org
). The organization claims more than 300 member companies from 20 different countries.
The IEEE and its sub-committees, which are working on tweaking specifications and ratifying the final standard, promise it will be backward compatible with products that are based on 802.11n Draft 2.0. This basically means they will play well with the inner pipework of 802.11 a/b/g networks by incorporating the same legacy techniques as Orthogonal Frequency Division Multiplexing (OFDM), forward error correction (FEC) coding and interleaving.
Most experts agree that compatibility is no longer an issue with 802.11n. Instead, enterprise users should look at such things as how a particular product handles high-speed network traffic, how much traffic management flexibility is provided at the edge points of the wireless network, and the overall security of the 802.11n architecture. Not all 802.11n products are exactly alike in terms of how the data is shuttled and passed along through the network.
"Having the ability to leverage and adjust this architecture is a competitive advantage," says Ken Lynch, a spokesman for Bluesocket, which has taken a strong stand on 802.11n and is expected to soon introduce an 802.11n product.
"There are going to be more bandwidth-intensive applications, and voice applications that require low latency and low jitter on the network" he adds. "It is important to have the ability to segment the network and decide what traffic on the network will route back to the controller, and what will stay resident locally and be more of an [access point-to-access point] communication."
Despite the technology's ability to improve the performance of current networks, Bluesocket supports a gradual move to 802.11n now rather than a full-front assault. "When high bandwidth applications become more prevalent, this architecture can support them well," says Lynch.
A number of wireless solutions providers are positioning 802.11n as a cost-effective alternative to wired Ethernet systems, since it offers broadband speeds and wider, more reliable coverage than conventional 802.11x networks.
This makes sense, since Ethernet cable is not only expensive, but it can be challenging to string it through older buildings and via already crowded conduits. Once installed, wired Ethernet is pretty much there to stay, since it usually cannot be removed without much difficulty or expense.
802.11n marks the beginning of a rapid market shift away from LAN access deployments using traditional wired Ethernet, says Paul DeBeasi, senior analyst for Burton Group. "The definitive and unalterable competitive advantage that 802.11n has over Ethernet is pervasive mobility," he says in a new Burton Group report.
The Burton Group predicts that 802.11n wireless technology will start to erode the wired Ethernet market within the next 24-36 months.
So, when should enterprise users take a serious look at 802.11n? Any or all of the following criteria are reasons to get busy:
* When laptops outnumber desktops in an organization;
* When mobile applications are the norm;
* When Fast Ethernet throughput just doesn't cut the mustard any longer.
* When networks are nearing the end of their lifespans (estimated at three to five years for a typical 802.11 network and five years for a wired system).
An Aberdeen Research study commissioned by a handful of wireless vendors echoes this forecast. More than eight in 10 (82 percent) of the 315 organizations surveyed will deploy 802.11n networks within the next 24 months.
Meru Networks, which was one of the vendors backing the Aberdeen study, is one of several companies promoting higher-bandwidth 802.11n as a direct replacement for wired networks. The quality-of-service benefits, and the addition of multiple-input multiple out (MiMo) technology, result in a great performance boost, says Meru's Ahlawat. "End users are beginning to ask us if they should be looking at the edge of networks, and do we need to worry about wired upgrades or not?"
A CLASS ACT
For some end users, the immediate performance benefits and relative cost savings offered by 802.11n are too much to resist. Morrisville State College, a State University of New York (SUNY) campus located 30 miles southeast of Syracuse, is replacing its entire wireless network with 802.11n technology, positioning the school as one of the first "pure" next-generation Wi-Fi networks in the country.
The university started the transition over the summer, working with Meru Networks and IBM Global Technology Services to install 10 Meru 802.11n access points (APs) in food courts, campus coffee shops and other areas where students congregate. In October, the push increased substantially as the school worked towards its goal of deploying 700 802.11n APs before the end of the month.
Morrisville's 802.11n network replaces an aging Raylink wireless network that was installed in 1999 and was incompatible with newer operating environments such as Microsoft's Vista, says Jean Boland, VP of I.T. services at Morrisville State College. She admits that the decision to replace the entire network with 802.11n technology was driven by the fact that the school was facing a "forklift upgrade".
Boland says it's still too early to gather any real data measuring the effectiveness and performance of the new network. However, her staff has already noted a dramatic boost in coverage and signal strength using fewer APs (the previous network consisted of 300-400 Raylink APs). She has also learned that, as yet, there are no tools specifically designed to deploy and measure 802.11n network statistics. In fact, while the access points are 802.11n, the tools used by Meru and IBM to install them are related to the older 802.11a standards. This could potentially cause complications since the operational characteristics of 802.11 a/b/g and /n differ dramatically, so using the older tools may not be the best way to deploy 802.11n to get the best results.
Still, the benefits of using the latest in wireless technology far outweigh the present lack of tools and ROI measurement models, and aren't always related to the nuts and bolts of I.T. By providing wireless at every nook and cranny of the university, the faculty is better preparing its students for the realities of the business world, notes Boland. "Two thirds of the workforce [is] mobile and when they leave the college they need to know how to work in that mobile environment."
Tim Scannell is president of Shoreline Research and a managing director of 2in10's consultancies in Boston and Scotland.