Dual-Mode 802.11 Put to Test

 
 
By John Taschek  |  Posted 2002-11-04 Email Print this article Print
 
 
 
 
 
 
 

802.11a/b devices run as fast as—or faster than—their single-mode counterparts in eWeek benchmark.

Its dual time.

Companies that cant decide on 802.11a or 802.11b can have both. A line of dual-mode access points from D-Link Systems Inc. and dual-mode PC Cards from D-Link and Netgear Inc. give corporations added wireless flexibility. Theyre also designed well enough so that the performance of dual cards and access points is at least as good as their single-mode counterparts, if not better.

eWeek Labs tested an assortment of devices, including a $180 Netgear 802.11a/b PC Card and the $129 D-Link DWL-650 802.11a card (both of which shipped last month); a single-mode Netgear HR-314 802.11a access point (which shipped in late August and retails for less than $400); and the $399 D-Link AirPro DI-764 dual-mode access point (which shipped last month). We also used an 802.11b device integrated into an IBM ThinkPad X30 for a base-line comparison.

While 802.11b is becoming pervasive and is integrated into a large number of notebooks and other devices (such as the Toshiba Corp. e740 Pocket PC and many more portable devices), support for 802.11a is coming on strong as well.

In the enterprise, 802.11as performance is very appealing, since the total bandwidth is shared among all users. This means that corporations can install fewer access points in dense environments while actually increasing overall user performance.

In the past, wireless cards were either 802.11a or 802.11b. Recently, Netgear, D-Link and others launched dual-mode PC Card solutions that support both standards. This means that each card contains two transceivers in the same-size form factor.

We highly recommend dual cards to all but those in controlled environments in which the configuration will not change.

Performance variations among brands of access points are generally minimal, although the way the circuitry is laid out can make for some differences. Likewise, performance differences among network cards are also minimal, probably because almost all 802.11a cards are based on the same Atheros Communications Inc. technology.

However, 802.11a performance remains at least two times faster than 802.11b devices, although its clear that distance creates drastically diminishing returns.

Although the performance of the D-Link access points is better over distances than that of Netgears equipment, we still recommend Netgear access points for most organizations. The company is one of the last that still uses metal encasements, which are more suited for placement above the ceiling in the plenum. They also simply look and feel more corporate.

However, small companies and departments should consider D-Links devices because they have an attractive price/performance ratio.

In tests, we used NetIQ Inc.s Chariot software to simulate database, e-mail and file transfer traffic. We combined a database test—a complex benchmark that sends transactions and waits for the update and a confirmation—with IBMs Lotus Software divisions Notes send-and-receive test, and we weighted that as 40 percent of the benchmark score. We then ran a series of combined file-based tests, which send and receive 100KB files, as 60 percent of the score.

In general, file-based testing shows a systems absolute maximum performance, whereas database and e-mail tests demonstrate how well a device handles mixed loads.

As expected, the D-Link DWL-650 802.11a PC Cards outperformed the 802.11b wireless solution integrated into the IBM ThinkPad. We expected that 802.11 cards would be underachievers when compared with what the performance specification allows (54M bps for 802.11a versus 11M bps for 802.11b). The DWL-650 cards were only three times faster than the 802.11b IBM solution at 30 feet.

That delta narrowed considerably when we stretched the test range from 30 feet to 100 feet in a multicubicle office. The DWL-650 performed only 181 percent faster than the 802.11b system, even after we optimized the positioning of the device, which is often difficult to do because there is a fixed antenna on the cards.

Were speculating that the reason for diminishing performance over distance is the card itself. The IBM ThinkPad has an integrated antenna thats better laid out than the PC Card antennas that jut out of the PCMCIA slot on most notebooks. In addition, integrated antennas such as those on the ThinkPad are less susceptible to interference, such as from people walking by or by positioning of hands or objects over the PC Card antenna.

Turbocharged

We also tested the turbo mode on the D-Link access point. Turbo mode reduces the number of channels available and is not suited for organizations with spanning access point configurations. Turbo mode is also only about 25 percent faster than non-Turbo mode in practice. (Marketing hype says it should be nearly twice as fast.) We recommend Turbo mode for home users and organizations with specific, limited needs.

802.11b gear in Turbo mode also provides gains. The 802.11b Turbo, released by Texas Instruments Inc. in the form of the ACX100 chip set, supposedly offers 22M-bps performance and a 70 percent greater coverage area (and a 30 percent linear coverage increase).

Initial tests show that Turbo mode on 802.11b yields a little less than a 30 percent increase in performance. We also noticed that products based on the TI design (basically, all the 802.11b products with "Turbo" in the name) do have a greater coverage area. For example, at 100 feet most of the TI-based units were operating at full speed, while other designs operated at slightly reduced speeds.

However, although the Turbo mode in 802.11b systems yield a significant gain, we recommend 802.11b Turbo for home networking only, since it is a proprietary solution based on TIs Packet Binary Convolutional Code technology, and there are no guarantees it will work with other cards, even when they are based on the TI chip set.

D-Link, meanwhile, has also released a dual-mode access point. We found that this solution performs as advertised. In fact, it performs marginally better at 100 feet than the single-mode Netgear 802.11a access point router. However, since this device is an either/or proposition—it operates at either 802.11a or 802.11b, not both—theres not much reason to opt for it unless corporations want to standardize on a companys product line and want a consistent management interface.

eWeek Labs Director John Taschek is at john_taschek@ziffdavis.com.

 
 
 
 
As the director of eWEEK Labs, John manages a staff that tests and analyzes a wide range of corporate technology products. He has been instrumental in expanding eWEEK Labs' analyses into actual user environments, and has continually engineered the Labs for accurate portrayal of true enterprise infrastructures. John also writes eWEEK's 'Wide Angle' column, which challenges readers interested in enterprise products and strategies to reconsider old assumptions and think about existing IT problems in new ways. Prior to his tenure at eWEEK, which started in 1994, Taschek headed up the performance testing lab at PC/Computing magazine (now called Smart Business). Taschek got his start in IT in Washington D.C., holding various technical positions at the National Alliance of Business and the Department of Housing and Urban Development. There, he and his colleagues assisted the government office with integrating the Windows desktop operating system with HUD's legacy mainframe and mid-range servers.
 
 
 
 
 
 
 

Submit a Comment

Loading Comments...
 
Manage your Newsletters: Login   Register My Newsletters























 
 
 
 
 
 
 
 
 
 
 
Rocket Fuel