Gary, I may have taken liberty in the use of the term symbiosis. But my point was clear enough. I do maintain, however, that the new strains of optical networking are indeed presenting a challenge on some level to traditional router and switch vendors.
Consider, the WAN is moving into the direction of flattened Ethernets and optically-driven MPLS-like link- and path- switching schemes. In due time we will begin seeing an increase in the number of edge-to-edge Layer 2 link setups, on the fly, carrying hundreds of scores of like-traffic payloads from as many end users, replacing the traditional up to 12 (or more) inter-networking journey "hops" that usually characterizes a "single" user's data session today. No doubt, Cisco was at the forefront of this approach with its Tag, but the state of the art is going far beyond that today.
The large router shops, too, have addressed this new double-edged "opportunity" with acquisitions of capable startups (e.g., Monterey, in CSCOs' case <I was happy to finally read a release on this box the other day, btw>), but that doesn't mean that their traditional router products wont be impacted by the others who are already ahead here. They will, and are. Sure, even these slower devices operating solely at Layer 3 will continue to thrive due to the overpowering tsunami of 'Net traffic, but not to the same extent as they would, otherwise.
When a clear direction is finally established -- i.e., once all of the conflicting initiatives and standards are sorted out and everyone agrees to play nice together, either under ITU or IETF or mutual agreements -- and once the arrow-takers fall by the wayside and the next routing regimen takes an unambiguous hold, then the top three players with a major stake in what's going on will quickly do what they have to, in order to ensure that they remain the top one, two and three.
No one here would suggest that it would be CSCO to "lead" the charge to supplant their existing 7xxx and 12xxx lines of routers. Not to mention all of the lower order models used for feeders and local distribution. Of course not. But they will attempt to stay abreast of disrputive opticals, or even attempt to stay ahead of many of the others, in order to hedge their interests.
But they still want to get as much mileage out of those existing devices as they possibly can, I would imagine, before the next wave of networking modalities moves into place. And I can't say that I blame them for that.
At the end of the day it's not so much what the large vendors want or don't want, as much as it is the same old story: Legacy Entrapment. Routed architectures have become the next legacy issue that leading edge deployers must contend with.
How do you put in the next better-cheaper-faster light-switch box when you still three or four years of the last thing you spent millions on sitting on the books? A bigger problem for carriers and SPs, perhaps, but end user organizations are facing this now, too.
And once again this will feed a form of inertia, just as today's Class 5s in the public voice space do, that the router folks will continue to feed off. This turns out to be a lot of discussions. Since I haven't any theological or vested financial leanings in any of this, I'll give it a rest. Just exercising my thoughts on the matter, and doing a bit of mind-blasting, if you will. Thanks for hanging in there.
FAC |
