re: 40 GigE Could End Standards Spat"... a combined standard is necessary because of problems with the current 10-Gigabit Ethernet standard. "The IEEE came out with 10-Gigabit Ethernet and it was adopted poorly because it was envisioned poorly."
I remember confusing discussions about which optical form factor, LAN-PHY vs WAN-PHY.
Exactly what was the problem the person referred to here?
re: 40 GigE Could End Standards SpatA few observations from the past:
1) Financial & Corporate politics holds much sway over technology decisions...did we really need multiple 100Base-T standards plus VGBase-AnyLan?
2) What was the real reason behind the 10GbE 850nm specification...could it be that somebody wanted to introduce a new fiber standard rather than find a way to make mid-priced transcievers work on the existing installed FDDI grade spec optical cable?
3) Why did the vendors come up with all the different 10GbE LAN and WAN specifications/standards? Good thing there were so many to choose from, right?
4) Why would the networking vendors want to adopt an existing 40GbE standard when they could force the market into a new 100GbE specification?
5) Why would the vendors who currently manufacture the existing 40GbE products be reluctant to abandon their existing investments?
Say what you want about technical aspects of one solution over the other...it all comes down to who wants ultimate control over a potentially large market...follow the money trail and you'll find why the split between the two camps exists.
Existing telco-focused manufacturers will prefer 40GbE. Networking focused vendors will push for 100GbE.
Seems this is a redux of the 10GbE LAN/WAN arguements.
In the end, it's the end-user who will suffer from more expensive than needed equipment, too many product choices, and a splintered market place until basic economics squeezes out a solution.
re: 40 GigE Could End Standards Spat"Who built the wavelength routing boxes, i see only books written by highly talented guys but no product i am aware of who can route wavelengths? Pls let me know if u know."
I seem to remember in some Lucent (now Alcatel-Lucent) labs a wavelength router. It was gathering dust but anyway apparently it was "functional".
re: 40 GigE Could End Standards Spat"Remember wavelength routing...it was built and nobody came"
Who built the wavelength routing boxes, i see only books written by highly talented guys but no product i am aware of who can route wavelengths? Pls let me know if u know.
"But yes someday it will be rebuilt and all would join"
I am also waiting for that someday, u have idea how far it is 5, 10, 15 years ..
re: 40 GigE Could End Standards Spat" seem to remember in some Lucent (now Alcatel-Lucent) labs a wavelength router. It was gathering dust but anyway apparently it was "functional"."
I believe the Lucent LambdaRouter was actually a switch.
re: 40 GigE Could End Standards SpatI seem to remember in some Lucent (now Alcatel-Lucent) labs a wavelength router. It was gathering dust but anyway apparently it was "functional".
i don't think that is gathering dust as far as i know there was no such product only experiments. If i try to search it take me to 2000 era where every marketing manager in those organizations was thinking that there will be thousands of wavelength running in the backbone. there was talks of 1024x1024 optical switch, optical muxes, etc. I read in 2000 Nortel bought a company called Xros going to build a 1024x1024 optical switch in 3.2 Billion. That time that company has around 32 employees. Every employee was worth $100 Million. I dont know what Nortel got but those employees got what they wanted. I know few companies who sell those kind of boxes are having 8x8 switch, do u see the difference in numbers they were talking in 2000 & 2007.
Do u read those kind of stuff now.No because they are not able to provide a reasonable high speed to the end user who can generate traffic of hundreds of wavelength.
re: 40 GigE Could End Standards SpatIn 1999, Monterey Networks built a Wavelength Router, then got acquired by Cisco, and then disappeared.... http://findarticles.com/p/arti...
Cisco ONS 15900 talk about the same thing but it is in End of sales/Life stage. In 1999 era on the basis of lab experiment start up companies were busy is collecting money/acquired. Now when i am searching wavelngth router in google, can't find single product. One more thing as why they named as router also does not make sense( already somebody has pointed out). What i read that time that suppose if there are N wavelength comming and N going out at a junction. It can switch wavelength from any incoming port to any outgoing port. JUst demultiplexing, switching and multiplexing again. There may be some more functions like power equalizer, dispersion equalizer etc.
I am not a switch/router guru but don't know whether it can be called router.
Monterey is talking about 160 Terabits, i don't know any vendor who has such a large Optical switch till today.
I remember confusing discussions about which optical form factor, LAN-PHY vs WAN-PHY.
Exactly what was the problem the person referred to here?
1) Financial & Corporate politics holds much sway over technology decisions...did we really need multiple 100Base-T standards plus VGBase-AnyLan?
2) What was the real reason behind the 10GbE 850nm specification...could it be that somebody wanted to introduce a new fiber standard rather than find a way to make mid-priced transcievers work on the existing installed FDDI grade spec optical cable?
3) Why did the vendors come up with all the different 10GbE LAN and WAN specifications/standards? Good thing there were so many to choose from, right?
4) Why would the networking vendors want to adopt an existing 40GbE standard when they could force the market into a new 100GbE specification?
5) Why would the vendors who currently manufacture the existing 40GbE products be reluctant to abandon their existing investments?
Say what you want about technical aspects of one solution over the other...it all comes down to who wants ultimate control over a potentially large market...follow the money trail and you'll find why the split between the two camps exists.
Existing telco-focused manufacturers will prefer 40GbE. Networking focused vendors will push for 100GbE.
Seems this is a redux of the 10GbE LAN/WAN arguements.
In the end, it's the end-user who will suffer from more expensive than needed equipment, too many product choices, and a splintered market place until basic economics squeezes out a solution.
I seem to remember in some Lucent (now Alcatel-Lucent) labs a wavelength router. It was gathering dust but anyway apparently it was "functional".
Who built the wavelength routing boxes, i see only books written by highly talented guys but no product i am aware of who can route wavelengths? Pls let me know if u know.
"But yes someday it will be rebuilt and all would join"
I am also waiting for that someday, u have idea how far it is 5, 10, 15 years ..
I believe the Lucent LambdaRouter was actually a switch.
i don't think that is gathering dust as far as i know there was no such product only experiments. If i try to search it take me to 2000 era where every marketing manager in those organizations was thinking that there will be thousands of wavelength running in the backbone. there was talks of 1024x1024 optical switch, optical muxes, etc. I read in 2000 Nortel bought a company called Xros going to build a 1024x1024 optical switch in 3.2 Billion. That time that company has around 32 employees. Every employee was worth $100 Million. I dont know what Nortel got but those employees got what they wanted. I know few companies who sell those kind of boxes are having 8x8 switch, do u see the difference in numbers they were talking in 2000 & 2007.
Do u read those kind of stuff now.No because they are not able to provide a reasonable high speed to the end user who can generate traffic of hundreds of wavelength.
http://findarticles.com/p/arti...
Cisco ONS 15900 talk about the same thing but it is in End of sales/Life stage.
In 1999 era on the basis of lab experiment start up companies were busy is collecting money/acquired. Now when i am searching wavelngth router in google, can't find single product.
One more thing as why they named as router also does not make sense( already somebody has pointed out). What i read that time that suppose if there are N wavelength comming and N going out at a junction. It can switch wavelength from any incoming port to any outgoing port. JUst demultiplexing, switching and multiplexing again. There may be some more functions like power equalizer, dispersion equalizer etc.
I am not a switch/router guru but don't know whether it can be called router.
Monterey is talking about 160 Terabits, i don't know any vendor who has such a large Optical switch till today.