Silicon Investor (SI) -- The First Internet Community

STOCKTALK

We've detected that you're using an ad content blocking browser plug-in or feature. Ads provide a critical source of revenue to the continued operation of Silicon Investor. We ask that you disable ad blocking while on Silicon Investor in the best interests of our community. If you are not using an ad blocker but are still receiving this message, make sure your browser's tracking protection is set to the 'standard' level.

Technology Stocks : Frank Coluccio Technology Forum - ASAP -- Ignore unavailable to you. Want to Upgrade?

To: pat mudge who wrote (1328)	4/7/2000 3:50:00 PM
From: Frank A. Coluccio	Read Replies (1) \| Respond to of 1782

Hello Pat, good to see you here. Re: Brightlink, nee Corvia: I think that I'd much rather see some white papering from BrightLink's own cadre of professionals than from Passmore and Flanigan (although I do happen to respect both of those individuals, as well). But even if we follow the logic in the netreference white paper, switched links in support of IP flows should not appear to be as radical as the writing would suggest. Writers often like to exaggerate the opposing qualities of these protocols instead of their prevailing symbiotic qualities. The writers tend to confuse the petty philosophical differences which tend to reflect territorial or NIH qualities between professionals who align themselves in manners consistent with the all-too-familiar bellhead-nethead face-off model, instead of realizing how truly interdependent or cohabitive the layers have become to one another in so many ways. And when they are not fully cohabitive, they sometimes tend mimic one another. By now we've already seen other vendors' products whose architectures use switched modes at the core of optical networks, either in the way of full lambda switching, or in the way of wavelength routing (a la Monterey's WaRP), in ways which mimic both ATM virtual path switching and IP's MPLS (which MPLS, itself, mimics ATM). Actually, IMO it's a misnomer to call lambda switched services "circtui switching" if you want to be entirely respectful of the lexicon... but what is physical today, and what is virtual, tends to be subjective, even vendor-specific. I have to take some pause here, however, and step back a moment to examine what it is, exactly, that these bright folks mean when they say they are avoiding the crosspoint model in favor of one that is hyper-toroidal (hypertorus) in nature. Sounds fascinating to me that a switching grid topoply defined in four-dimensional space, dependent on fractal principles, would be used to support their claims as to why they can do it better, but at the same time nowhere do they describe how or what this means, or why it should be believed. I therefore cannot comment one way or the other on their claims at this time, nor could I until I see more of what they have under the covers. BEGIN SIDEBAR: The white paper in question here deals with optical circuit switching. It is precisely what came to my mind yesterday when I came across the SmartPipes post here. But much to my surprise, the folks at SmartPipes --instead of taking a simplistic approach to shunting bulk bandwidth across domains-- have elected to take what has been described as "... a package of extremely complex software containing zillions of instructions that facilitate communication between all manner of software and hardware." hmm... To take the more complex approach in this manner is to deny that at some point in the future there will be an abundance of bandwidth. It is to suggest that SmartPipe's is a minimalist's approach to networking, in other words. At the same time it will in many cases unnecessarily perpetuate a dependence on the protocols of minutiae, for every packet sent. Certainly, in a focused enterprise network setting --where much greater controls are possible than, say, on the Open Internet-- it makes more sense at some point to switch like payloads in bulk, and if critical mass exists to join pipes of a like nature at the physical layer like a steamfitting, instead of administering to every packet individually, end to end. END SIDEBAR Getting back to BrightLink, I think we all need to see more substantive information written by the company itself. If you have other references, please share them with us. Thanks. Of course, if I've misunderstood, or if I am mistaken about anything I've written here concerning either SmartPipes or BrightLink, please advise. All comments and corrections welcome. FAC