re: Cisco, Sycamore Circling Lucent's ATMWhat is the market for ATM switches going forward. In 2001 and 2002 any one had dollar figures for the ATM market? out of this how much is new deployment?
Anybody has pointers to installation data?
Why would RBOCs or anyone else build data back bone using ATM going forward? 53-byte cell is a wrong decision made at a time when switching and semiconduactor technology is at least 2 times behind current state of the art. At least 70-90 byte cell might have given it longevity and made MPLS look unnecessary.
re: Cisco, Sycamore Circling Lucent's ATMBuying Lucents ATM boxes is a laughable idea. I think people will have realized by now (they better have!) that for the whole to be greater than the sum of parts you have to merge with good ideas and committed people who can *continue* developing good ideas, not just purchase varieties of (soon to be obsolete, ill maintained) god boxes.
All the good people have long since left Lucent. No one is going to pay 600 million dollars for lucents atm business.
re: Cisco, Sycamore Circling Lucent's ATMInstead of trying to buy ATM gear, Cisco should try and get their core router working.
I would like some clarification on my test scenario. I tried sending traffic through OC-192 interface on the 12416 and looped it back through the same interface, after verifying that traffic transits the switch fabric. The performance is just awful!!
At 40B, throughput is 100% but falls to ~ 67% at 41B and delay jumps to 10 milliseconds. Remember this is just one OC-192 interface. I checked the firmware version on the card, everything seems right. I can't believe that Cisco has been trying to sell this crap for this long!
re: Cisco, Sycamore Circling Lucent's ATMOne market for ATM that will not go away is the Federal Government. One big reason is that much of the encryption technology that is used by DOD and other agencies is designed for 53 byte cells. Having a single cell size allows for faster and deeper encryption without slowing down traffic. If the government went away from ATM, the cost to replace encryption technology will be very high so dont look for it.
As for Telecom networks, ATM will be replaced when the carriers do not make money on it. Right now, their Frame Relay networks are based on ATM and they are cash-positive money makers. Also, their people are trained to provision these systems. Why replace what works and incur the cost? So you see carriers augmenting their ATM on an "as needed" basis and reaping the profits. Something their IP networks have yet to do in spite of hundreds of millions of dollars in investment. Want to know what they will spend money on? Follow what they MAKE money on ....
re: Cisco, Sycamore Circling Lucent's ATMHi, The insane goal of supporting voice over ATM by 53 bytes cell is smart. In case, you did not know voice still represent the largest piece of revenue for service provider.. And, 60% of media gateways sold to service provider are Voice over ATM media gateway..
re: Cisco, Sycamore Circling Lucent's ATM> Why would RBOCs or anyone else build data back > bone using ATM going forward?
There are no alternatives at the moments.
> 53-byte cell is a > wrong decision made at a time when switching > and semiconduactor technology is at least 2 > times behind current state of the art. At least > 70-90 byte cell might have given it longevity
The insane payload size originated from insane goal of supporting ATM phones, henve there was a requirement to effectively map a single phone call to an ATM VC. It had nothing to do with semiconductors.
re: Cisco, Sycamore Circling Lucent's ATMorange: But if Dan puts up a help-wanted sign out front, most of us would be back. _________________________________________________
What is it that makes you go back to work for Dan and Desh? Just curious, coz I've heard strong opinions to the contrary.
re: Cisco, Sycamore Circling Lucent's ATMNot to mention the VoIP packet tax can be up to %50 when you get down to the small IP packet sizes that you need to keep latency down - VoATM starts looking more efficient.
Have to agree with Bobby on many of the points he made this time...