|E-mail Wes Simonds|
You say potato
I say potahto
You say tomato
I say tomahto
Let?s call the whole thing off.
Gershwin wasn't thinking of the complex task of the network manager in the twenty-first century when he wrote that, but the lyric still applies in some ways.
Network managers need X, fast, but only Y is actually on the market. They're promised X, but Y is what their carriers/ISPs wind up delivering. Sometimes it's not even Y; it's omega, which, as everyone knows, has serious interoperability problems with X in the context of an actual word.
Few aspects of networking are as mission-critical to today's net manager as a fast, reliable link to carriers/ISPs -- certainly for data, and increasingly for voice as well. And it seems to me that broadband wireless technologies, which promise high throughput, incredibly fast deployment times, an increasing range of reliability and value-added services, are starting to look sweeter and sweeter.
Oh, sure, you've heard the broadband wireless pitch before. And this is how you've responded: How appealing could it really be? Haven't we already solved those problems? The bottom line is that nothing is as fast or reliable as fiber, right?
That's certainly true for the most part. Binary data and fiber are a match made in heaven, and have been ever since they started making googly eyes at each other as teenagers way back in the day.
But fiber's expensive to roll out -- the cost having been estimated as high as $200,000 per square mile in some urban environments where it isn't already deployed. And surprisingly, more than 90% of US office buildings still haven't got a direct fiber connection.
Then, too, despite the gloom-and-doom economy discouraging cash expenditures, the fact remains that many businesses find themselves confronting situations in which they need today's fiber performance at tomorrow's delivery times.
Enter broadband wireless. By way of radio waves (generally referred to as fixed wireless technology) or lasers (known as free-space optics) transmitted in either a point-to-point or point-to-multipoint configuration, network managers can go from nothing to Gigabit Ethernet-class performance in literally a few days.
This is obviously a strong selling point in areas where fiber isn't already distributed like hurled spaghetti in every direction. No infrastructure? No problem. In Western Pennsylvania, for example, Prime Companies has been selling data services via fixed wireless for a solid year now.
But even in densely populated urban areas, a good business case often exists for the technology. In the weeks following Sept. 11, for instance, fiber services in Manhattan proved less than entirely reliable for reasons beyond anyone's control, and Merrill Lynch, which had considerable voice and data needs in the immediate term, had to find an alternate approach. Immediately.
Their solution? Terabeam, a Seattle-based free-optics solution provider. In a matter of days Terabeam had established a high-speed link from lower Manhattan to New Jersey -- over a mile -- which served Merrill Lynch's needs at a time when traditional approaches just weren't delivering.
Imagine that. You say potato, you get potato.
Naturally, both fixed wireless and free-space optics technologies have the obvious shortcoming: blockage is an issue for them, particularly weather blockage, whereas it's virtually nonexistent for fiber. Both technologies don?t get along well with dense fog, for instance, which from the standpoint of data throughput is an infinite number of water-based mirrors, which can reflect, distort, or even completely stall a transmission.
But bear in mind that Terabeam is based in Seattle and you begin to see exactly how much bad weather is really required to rain on the free-space optics data parade.
And the story is much the same for fixed wireless, which has become more and more robust with each generation and which is looking increasingly appealing as a cheaper alternative to business DSL in point-to-multipoint configurations in cities.
You know, I hate to say it, but despite the title of this column, I'm having a tough time finding anything bad to say about these packets. The fact is that I think broadband wireless is a great networking concept whose time has come.
Next time around, though, I'm writing about Microsoft.
Bad Packets: The network manager's line between Us and Them