The labyrinthine process of calculating a WLAN TCO

Currently, enterprise-class wireless LAN access point prices are all over the map, due to a number of factors. Among them:

* How much, if any, software intelligence is in the device.

* How much custom development must be done on the part of the vendor for value-added features.

* Whether special processors are required to provide the horsepower to execute many of the software features, such as encryption.

* In what volumes the devices are produced.

Of course, the AP is only one component of your entire total cost of ownership (TCO) calculation. I agree to a certain point with Chris Kozup, program director, infrastructure strategies at the Meta Group, who suggests that you ignore marketing-speak like “lightweight APs” and “WLAN switches.”

Rather, look at the total system being offered. See if it meets your feature requirements and whether it jibes with your religious philosophy about where to house certain functions like QoS and security. Then evaluate TCO both on capital costs and the ongoing operational savings you can glean based on management tools and features offered.

Otherwise, the whole landscape of terminology and architecture twists and turns can be paralyzing.

“Many customers are postponing procurements because they can’t figure out what’s best to deploy,” Kozup says.

Let’s look again at just the AP component.

Symbol Technologies was officially the first to introduce the concept of a scaled down AP, with its network services and management capabilities offloaded into a controller, or “switch” device that sits either in a wiring closet or data center. One big goal was to centralize (and thus ease and lower the expense of) securing and managing large numbers of APs.

Another goal was to keep the capital cost of APs down. This variable becomes more significant as their volume increases. 

Let’s say you only need four 802.11b APs. The difference between $249 (Symbol radios) and $679 (Trapeze radios) doesn’t make a big difference to the bottom line. You’d spend $1,720 more on the price of the radios using Trapeze ($2,716) than Symbol ($996). In the grand scheme of life, $1,720 isn’t going to make a huge difference.

But let’s say you need to grow your implementation to 100 radios. The Symbol AP component cost is $24,900 compared to the Trapeze AP cost of $67,900. Now the capital cost differential is noticeable ($43,000).

The grander the scale of the implementation, the more APs you need to buy. But, also, the more robust and scalable you need your centralized services and management features. At this point, the capital costs of the APs need to be considered alongside the price of a switch and how many APs it can support, the network service feature set, whether you need to buy a separate device to acquire power over Ethernet and that device’s cost, RF management tools, and so forth.

This all being said, is the “lightweight” model all that significant? You can get a single-mode access point from Cisco running the full suite of IOS software with an 802.11b or .11g radio (no upgradability, though), including antenna, for $599 (this is the Aironet 1100 series, which supports just one radio).

And you don’t have to buy a separate management appliance for traditional monitoring and management (you can use SNMP managers such as HP OpenView).

If you want Cisco’s latest RF tools (which you probably do for a large installation), you do need the company’s Wireless LAN Solutions Engine 2.5. That price is $8,495. But it manages up to 2,500 access points. If you divide that out, that’s about an extra 3 bucks per AP for management if fully utilized.

There are more variables for your spreadsheet, of course, but this is food for thought.