Polatis Plugs On With Optical Switches

COMMENTS

Newest First | Oldest First | Threaded View

<< < Page 3 / 5 > >>

o-man 12/5/2012 | 1:23:28 AM

re: Polatis Plugs On With Optical Switches Balet,

The transponder whore is a great topic... my 2 cents:

1. Optium
2. CivCom
3. Agility (rumor of failure to deliver tunables to a tier 1)
4. Oplink
5. JDSU
6. AVANEX

What do you think?

Reply | Post Message | MESSAGES LIST | START A BOARD

o-man 12/5/2012 | 1:23:27 AM

re: Polatis Plugs On With Optical Switches nice point raymand!

in the end it ALWAYS is reduced to price. I think cheap hookers are used more that the expensive ones :)

Reply | Post Message | MESSAGES LIST | START A BOARD

Balet 12/5/2012 | 1:23:26 AM

re: Polatis Plugs On With Optical Switches Very good questions, Raymond.
If anyone new answers today he/she would've been rich about 5 years from now.

I think the packet and burst switching is still too far away. I don't see any resonable optical technology to deal with packets fast enough and inexpensive enough. Do you?

Reply | Post Message | MESSAGES LIST | START A BOARD

Balet 12/5/2012 | 1:23:26 AM

re: Polatis Plugs On With Optical Switches You are right, optical. There is not enough MEMS switches built to come up with any resonable MTTF numbers.
On "old" opto-mechanical ones the reliability was extremely poor, especially Dicon and E-tek switches. Dicon was making more $$ from fixing them them from shipping them in 1999-2000.

There is a new breed of PLC switches coming to market that have a potential fro as good reliability as electronics chips.

Reliability could be a big hit for Polatis, by the way.

Reply | Post Message | MESSAGES LIST | START A BOARD

optical 12/5/2012 | 1:23:26 AM

re: Polatis Plugs On With Optical Switches I think one other issue that needs to be addressed (besides cost) is the reliability of OOO switches. Are 3D MEMS based switches really ready to handle the rigors of real life carrier class deployment? What are the MTBF numbers on these OOO switches? I'd be interested in hearing people's thoughts on this.

Reply | Post Message | MESSAGES LIST | START A BOARD

shaggy 12/5/2012 | 1:23:25 AM

re: Polatis Plugs On With Optical Switches O-man:

BTW does any one know the OEO from infinera??

..

It is the announcement of Infinera's prototype that led me to my original comment-if they can keep doing what they have always been doing, but 10X cheaper, what motivation have they (optical network operators) to change architectures?

Still much risk re viability at Infinera. This integrated optics and silicon concept has been run up the pole a few times and failed. Given the team, however, I give them a much better chance of success.

Reply | Post Message | MESSAGES LIST | START A BOARD

o-man 12/5/2012 | 1:23:24 AM

re: Polatis Plugs On With Optical Switches Was Kotura doing silicon optics? The last I heard they got rid of their CEO and VP of BD&Marketing.

I hope the team at Infinera is better than BB!!

Reply | Post Message | MESSAGES LIST | START A BOARD

shaggy 12/5/2012 | 1:23:24 AM

re: Polatis Plugs On With Optical Switches O-man:
"in the end it ALWAYS is reduced to price. I think cheap hookers are used more that the expensive ones :)"

I will leave it up to you regarding hooker use analysis, however, you raise a valid point as it relates to price- carriers care little for the technology used to achieve a desired functionality, so long as it meets cost targets, is reliable, and can be readily produced (cost again).

Reply | Post Message | MESSAGES LIST | START A BOARD

shaggy 12/5/2012 | 1:23:24 AM

re: Polatis Plugs On With Optical Switches Raymand:

"All optical wavelength/circuit switching would seem to be losing its advantages as networks are rapidly becoming data centric. I assume that data centric networks benefit more from distributed fine grained routing than from static large bandwidth wavelength architectures."

But....

data-centric networks also lend themselves to mesh/point-to-point network architectures, as IP is all hop-to-hop anyway. If the value of a dynamically controlled network could be demonstrated, all-optical switching could play a role here.

I see more viability for all-optical switches in dynamic wavelength switch applications as the networks become more data-centric. That said, if you can put 80 OE terminations on a single chip, why bother?

Reply | Post Message | MESSAGES LIST | START A BOARD

romulusx 12/5/2012 | 1:23:22 AM

re: Polatis Plugs On With Optical Switches Hybrid OEO/OOO switches are not just a science project. Corvis has been selling such a swith to the US Government for several quarters now. As far as "All Optical" networking - the closest thing to that would be Broadwing (which Corvis owns). Dr. Huber (Corvis CEO) architecture is built on the premise that electronics cannot be eliminated in the Metro space at this time. Corvis was created to address what he felt would be the ultimate need for an OOO architecture - that being the long haul arena. He estimates that 80% of all traffic is long haul. If you need any further validation of this architecture take a look at Broadwing Connect family of products. This line of products has been praised for its ability to offer benefits that typical Sonet networks just can't offer.

Reply | Post Message | MESSAGES LIST | START A BOARD

<< < Page 3 / 5 > >>