Courtesy of Karl Denninger at The Market Ticker
Cisco CRS-3, powered by Cisco QuantumFlow Array – a chipset architecture engineered in multiple dimensions of scale, services, and savings.
That’s the announcement. It was the cause of all the hype. A "new dimension" that works within their existing CRS framework.
Basically, a faster version of the CRS-1.
The San Jose, Calif.-based Cisco had sent out invitations to analysts and the media for a "significant announcement" that it says "will forever change the Internet and its impact on consumers, businesses and governments."
You’d think there was some new technology. Something that nobody had seen before. Something revolutionary.
You would be dead wrong.
Now don’t get me wrong – more speed is good, of course. More capacity is good. More density – an increasing problem for various network folks, is never a bad thing, although there’s no such thing as a free lunch – more capacity in a smaller space comes with higher power requirements and heat loading, which in turn invalidates assumptions made by carriers, ISPs and others on the adequacy of power and cooling systems in their machine rooms – sometimes with extremely expensive consequences.
But "forever change the Internet and its impact on consumers, businesses and governments"?
This reminds me of the hype when the CISCO 7xxx carrier series routes were introduced in the 1990s. These were "going to change the Internet forever" too.
But that was a forced upgrade CISCO was able to capitalize on due to their near-monopoly position in the core routing space at the time.
What was only known to people who understood the Internet at the time (myself included) was that the reason that device garnered instant acceptance and huge order flow was that the Internet’s routing table was exceeding the storage capacity of the CISCO AGS+ which was, at that point, mostly at the core of the network. Carrier routers were literally crashing as a consequence of running out of memory, and the architecture of the AGS+, which was roughly-based on the VME backplane architecture and the Motorola 68xxx processor line, did not allow for expansion of the address space.
In other words CISCO had painted itself into a corner and its response was to come out with a new, improved architecture that didn’t have those limits – and the Internet’s address space expansion effectively forced every ISP in the United States to buy their 7xxx series gear almost immediately.
But this is a different matter entirely. There is no "forced corner" that CISCO can exploit here, there is no monopoly position in the marketplace as there was with the AGS+ 20ish years ago and while this is obviously a step forward in performance-per-unit-of-density the firm’s claims are, in my opinion,dramatically overblown.
I used to have respect for John Chambers. Not any more.
That John Chambers sees it necessary to resort to this sort of over-the-top hyperbole unfortunately causes me to have flashbacks of 1999 – when every "technology company" had their stock pumped day after day by releasing utterly overblown and outrageously hyped press releases, while the actual market impact of what made it to market was more akin to a mosquito biting an elephant’s ass.
Let’s hope what follows for those who buy into this bilge isn’t March of 2000.
Disclosure: No position.
PS: Would you mind disclosing the particulars and performance of the off-balance-sheet financing you’re carrying back for people – and the terms on that financing? (p63.) I’m just curious if there’s been some, uh, "deterioration" in the credit quality in that nice black box. Thanks in advance for what I expect will be a non-answer.