re: Chambers Expects Cisco Dominance>Cisco got one bonus announcement out of the >Calif., facility, said Kathryn Walker, Sprint's >executive vice president of network services. >Volpi, on stage at the time, said even he wasn't >aware this had happened.
If you followed the webcast, this one was clearly simulated and pre-planned.
re: Chambers Expects Cisco DominanceSo Chambers is a total stud. Love the guy. One of my heroes. No really. How could you not love this guy? He's like the poster child for Silicon Valley success. Thanks John! You give us cannon fodder for taking shots at Canadians. Just kidding, Canadians. You built my bikes. Love you too.
But this is a lame comment and a lame press tour. Reminds me of the movie "Gladiator" where the #2 guy in the Roman Army says "people should know when they're beaten." Russell Crow says "would you? would I?"
Cool scene.
Now John, what should the rest of us do: go run and hide? Should we throw in the towel because you say Cisco will run roughshod? Screw that.
Show me a new strategic product that's actually getting deployed successfully. If Juniper's 60 T Series customers haven't even seen this thing, are we to believe that it is anywhere close to deployment?
How about that 2-4k's. Out of gas and the Asians are coming, not to mention Dell. How about the Cat6K. Out of gas. How about WLAN switch. SWAN is the ugly duckling and you know it. Linksys? Fuggedaboutit. Crosses my home phone. Lame. How about carrier? What happened to your DSLAM business? How about optical? Are you a leader ANYWHERE in those cateogories? How about mobile infrastructure. Is the 7xxx really going to deliver IP services to my handset?
So how exactly does Cisco expect to dominate the world? When the customers have completely submitted to FUD and decided just to buy Cisco because it's Cisco. Didn't work for IBM and it ain't gonna work for Cisco.
re: Chambers Expects Cisco DominanceThe bigger the hype, the less the substance. When you invest 500M in new product line would be nice to have some customer wins under your belt vs. some lukewarm testimonials and trials. Had to rush this thing out given loss of core share for last 6 months to Juniper. Will carriers sit tight for another year until this thing is finished? Good luck. At least Cisco is innovating though.
re: Chambers Expects Cisco DominanceThe other thing that struck me was the idea that the new version of IOS could eventually percolate throughout Cisco's products, effectively ridding it of its biggest intrinsic problem....it's tangle of different versions of IOS running of different products.
The comparison with Microsoft's shift from DOS to Windows NT is exciting, but....
- how feasible is this?
- other aspects of this analogy aren't so appealing, like having software that makes very inefficient use of processor power, and having to upgrade hardware to use the latest Cisco IOS
re: Chambers Expects Cisco DominanceI was struck by Chambers' quote about the longevity of the new CRS-1:
"Once it's in place, we want it to be there one or two decades."
Do you think the development of the Internet is now predictable enough to make that assertion?
For instance:
1. Is the debate over where intelligence should reside in the network - whether we'll end up with a dumb optical core surrounded by edge routers - has now been settled?
2. Up until now, yesterday's core routers have often been repurposed as today's edge routers. Is this not going to happen in the future?
Yes, I believe that we can reasonably expect that routers will continue to be a big part of the Internet for years to come. I think part of your question regarding whether the optical core is dumb or smart is kind of irrelevant. Regardless of the answer, routers are going to be necessary unless we get a breakthrough in optical packet routing. I'm not holding my breath.
The whole point about repurposing core routers as edge routers is that it allowed the carriers to extend some of the useful life of the router and thus gain some more value while still in the depreciation window. The carriers would GREATLY prefer not to have to do this and instead also use such large routers as their edge routers. This would eliminate at least one layer and possibly two layers of devices in their network. The grand, long term vision is an entire POP driven by a single router that does core, aggregation and edge services.
The question you didn't ask is what does it take to fulfill the requirement of one or two decades. If we do just a wee bit of math, we get some very interesting numbers. Suppose that the Internet continues to grow at 100% Y/Y as it has for the last decade or so. In one decade, that turns out to be a growth factor of 1000X. For two decades, you need a growth factor of 1,000,000X.
Your article quotes John as expecting growth rates of 400-500% Y/Y. Those numbers give growth factors of 9,000,000X to a whopping 3x10^15. Clearly, these numbers are unsupportable, but even the more conservative 100% growth rate numbers would seem to indicate that we would need to replace every single OC-192 that we have today with an entire CRS-1 in the next ten years. And since folks already have multiple OC-192s into single routers today, the obvious conclusion is that the CRS-1 is still too small from an architectural maximum point of view.
The pragmatic reality is that Cisco is still working on the process of getting a single unified software image that runs on the 6500. Today, about a decade after acquiring Crescendo, CatOS and IOS Classic are still separate options. Porting IOX to the existing product line is just not going to happen in our lifetime.
Even getting IOX to replace IOS Classic will be a herculean task. Realize that the HFR implements a very restricted subset of IOS functionality, and that many things are not supported on the GSR either. Getting support of all of the Enterprise features into IOX would be a huge task.
Given these timeframes, I'd only expect to see IOX on new products for quite awhile.
The whole point about repurposing core routers as edge routers is that it allowed the carriers to extend some of the useful life of the router and thus gain some more value while still in the depreciation window.
This is what carriers have been doing since the dawn of time with all of their telecom gear.
The carriers would GREATLY prefer not to have to do this and instead also use such large routers as their edge routers. This would eliminate at least one layer and possibly two layers of devices in their network. The grand, long term vision is an entire POP driven by a single router that does core, aggregation and edge services.
I would put this completely differently: carriers EXPECT to be able to do this and PLAN for it because it is what they have always done. Obviously you and I have talked to very different carriers Tony. The ones that I have talked to have a vision of many providers' equipment (gives them price leverage) all working together nicely under the same OSS. Every carrier that I have ever talked to (includes Verizon, SBC, Bell Canada, MCI, Worldcom, Qwest, BT, and a few smaller ones) know that their POPs are not the same and therefore cannot use a single device. It just does not make economic sense.
If you followed the webcast, this one was clearly simulated and pre-planned.