DenverTechie,
I linked this to your response to Ken's post about Nortel. I believe what I have pasted below is somewhat related to Ken's questions. I would appreciate any general comments you have on our two articles concerning Northern Telecom. My post below is somewhat controversial, and I wish I knew more about the technology to be able to edit it, but I don't know enough and I was afraid it would lose it's meaning. It's not my personal comments. It's from another SI member.
Here's my basic concern (and the $64,000 question) with Nortel and, for that matter, any traditional TDM circuit switched telecom equipment supplier. I am concerned that they may not have the products needed to offer to their customers new packet switched data networking gear.
I remember reading the SEC filings on Newbridge Networks, and I was surprised how fast their traditional bread and butter TDM equipment sales are slowing and how rapidly their ATM based equipment sales are increasing. Not that I think that highly of NN, but only use them as an example. For some reason(?) I feel they are less prepared than other telecom equipment companies.
So my very general question is, in light of the information below, how does Nortel look to you during this transistion period from circuit switched to packet switched networks? Does Nortel/Bay have the products the telecoms, ISPs, and MSOs need to have for their network upgrades? In your opinion, are they really as weak in the ATM arena as the comments below suggest?
Thanks,
MikeM(From Florida)
_____________________________
Nortel IS very strong in circuit switching, BUT it is a Nortel created myth that they are strong in ATM. Nortel has three ATM products:
1. Passport family of switches
2. The Vector - a re-engineered FORE ASX-1000 switch and
3. The Concorde - Big Core ATM switch..
Here is where the myth comes into play:
1. The Passport-- (router/switch) is a multiservice box with 1.6 Gbps of throughput capacity (obviously too small to be a good ATM switch) Nortel sells a ton of these in Europe and the most popular interfaces are TDM/X.23/and Frame Relay (they have ATM interfaces but mostly used for uplinks to bigger ATM switches)..BUT because the internal architecture of the 1.6 Gbps backplane is cell based (ATM), Nortel counts the shipment of these switches towards ATM revenue and uses the shipment numbers as ATM products when they report to industry analysts. There are easily over 8,000 Passports (of varying size) installed around the world. The only carrier install in the U.S. is at Sprint for Frame Relay (and these switches may be removed and be replaced with Cisco products). Most U.S. Passport installs are in enterprise/corporate accounts. In Europe they are all over the place.
So, if you sell a switch loaded with Frame Relay modules, you can count it as both Frame Relay and ATM revenue and report it to analysts that way. It's nothing more than marketing B.S. Why can they get away with this? Well, because the internal architecture is ATM, every module has to have an ATM interface to communicate across the backplane...So, an eight port frame relay modules (8 external ports on the front) needs to have at least 1 ATM interface on the back to communicate to other modules in the chassis and in the case of the passport, each module as 2 internal ports because there are 2,800 Mbps busses on the backplane. It's a pretty lame accounting method, but a lot of vendors do this in-order to inflate their market share and revenue in fast growing technologies. This makes them more attractive to investors and customers who falsely believe that the company is on the right technology curve. When NN sells a 36170 ATM switch loaded with Frame Relay, what do you think they count it a...1. Frame Relay revenue, 2. ATM revenue or 3. Both? Which looks best to investors and industry analysts? Marketing B.S.
2. Nortel Vector-- A pure ATM switch. Nortel does nothing more than just slap their name on it. They totally re-enigneered it and built some redundacny features into it to make it more attractive to Service Providers/Carrier customers. Nortel sold between 125 and 175 units world wide - practically nothing. The problem the Vector doesn't sell well is because in the Carrier market, customers still see the Vector as a LAN ATM switch with some minor redundancy feature built into it...not a true carrier class product. In the enterprise accounts, corporate customers would rather buy the product from Fore because it's cheaper, the same thing, and if you are going to call a vendor for service issues...it's always better to call the manufacturer who actually created it, not a middleman at Nortel who would probably have to turn around and call Fore to get the issues answered.
The Nortel/Fore deal expired last January. They are going forward with the 'partnership' in name only. There are no more legal documents. Plus the Bay deal would have killed the 'partnership'.....
On a side note- Bay was OEM'ing NEC ATM switches (except for an old Centillion 100 Lan switch) nobody has commented on the future of the deal.. but you can bet it off. Nortel surely doesn't want to give NEC any sales revenue and NEC doesn't want to help Nortel - they are old PBX/Circuit switching enemies.
3. Concorde-- Supposed to be a core switch...it kind of is, when core ATM switches came on the market in 1993 there were really only 2 serious vendors..
1. ATT/Lucent with the Globeview 2000
2. Nortel with the Concorde
There were a handful of others like Fujitsu, Alcatel, Hitachi and Siemens. The Globeview was the market leader with 50% of the install base (I don't know the # of switches) and the Concorde has the second largest install base (I forget the %) BUT (the interesting thing) the total number of Concordes ever shipped (until last January) was 56. The larget single customer is a Chinese network with 8 units. So, Nortel can claim to be the 2nd largest Core ATM switch vendor in the world (but they have shipped less than 50 units).
More marketing than reality, Nortel's ATM market perception may be misleading. |
