Thread Links Date Links
Thread Prev Thread Next Thread Index Date Prev Date Next Date Index

Re: [HSSG] Reach Objectives: patch panel and WDM




Piers,
please clarify these sentences.

This might imply that just a few wavelengths per MAC-size might be less economic per unit of throughput than either 1 or more than a few (e.g. 8), when the full MAC-size.  In absolute terms, not buying the wavelengths you don't need might still be more economic.

They seem to be art odds with each other.  In your view, is the option to use a few of the total wavelengths more or less economic?

Regards,
Paul Kolesar
CommScope Enterprise® Solutions
1300 East Lookout Drive
Richardson, TX 75082
Phone:  972.792.3155
Fax:      972.792.3111
eMail:   pkolesar@commscope.com



"DAWE,PIERS" <piers.dawe@AVAGOTECH.COM>

08/23/2006 03:36 PM
Please respond to
"DAWE,PIERS" <piers.dawe@AVAGOTECH.COM>

To
STDS-802-3-HSSG@listserv.ieee.org
cc
Subject
Re: [HSSG] Reach Objectives: patch panel and WDM





If someone is organising his data center cost-effectively, he may be able to position his pair of new super-powerful routers so that a direct connection can be used, and save some money on patch panels.  After all, he couldn't patch the new routers to anything else!  This tight co-location would be even more strong in something like a multi-rack supercomputer (but I doubt that the backplane extenders of a supercomputer are really Ethernet, in market characteristics).
 
Also, with multiple lambdas there has to be a pair of wavelength multiplexers somewhere (even if hidden in a box or chip with the lasers), and these have an NRE cost and a marginal cost, in dollars and in power budget.  This might imply that just a few wavelengths per MAC-size might be less economic per unit of throughput than either 1 or more than a few (e.g. 8), when the full MAC-size.  In absolute terms, not buying the wavelengths you don't need might still be more economic.
 
Piers
-----Original Message-----
From:
Roger Merel [mailto:roger@LUXTERA.COM]
Sent:
22 August 2006 19:55
To:
STDS-802-3-HSSG@listserv.ieee.org
Subject:
Re: [HSSG] Reach Objectives

Agree with Matt that once you are considering a new cable (ribbon MMF), the considerations are different, and SMF make the distinction between 50m/100m/300m/2km irrelevant.  One PMD should address all of them.
 
However, the optimal solution may not be parallel SMF fibers but rather multiple-lambdas over a single-pair of SMF.  In which case you don’t have to suffer the cost of ribbon fiber (of either SMF or OM3) which can be less economically attractive both “per meter” and “per patch panel”.
 
 -Roger
 



From: Matt Traverso [mailto:matt.traverso@GMAIL.COM]
Sent:
Tuesday, August 22, 2006 1:47 PM
To:
STDS-802-3-HSSG@listserv.ieee.org
Subject:
Re: [HSSG] Reach Objectives

 
Hi Petar, John,

Since the OM3 fiber we are discussing would primarily be parallel, is there merit in defining  an OM3 PMD versus specifying a parallel SMF based PMD?  

My assumption is that in both cases the data center in question would be pulling new fiber.  If my assumption is true, then we should not limit our consideration to OM3,but should also look at SMF.  For example SMF would allow for the extension to greater distances more simply.

There may indeed be merit in defining the data center PMD to be OM3 based, but my opinion is that we should not jump to conclusions.

thanks
--Matt Traverso
Opnext, Inc.

mtraverso@opnext.com

Note - I sent this from a personal account to avoid an automated statement from my IT dept which contains phrases that would likely be considered "unfriendly" to the reflector.

On 8/22/06, Petar Pepeljugoski <petarp@us.ibm.com> wrote:

Brad,


I agree with you. And the breakpoints vary from solution to solution.


Let's consider the following two examples:


1. N ch. x 10 Gb/s over OM3 fiber at 50m, and at 100m, there is very little difference in the specs, so the cost will most likely not be affected.
2. M ch. x 17 Gb/s (or 20 gb/s) over OM3, there will be much larger difference between 50m and 100m and there might be cost premium.


This will depend on the expected yield distribution and how sensitive is to changes in the specs. So, we will need to carefully tread and not become prisoners to the reach objectives, but rather optimize the cost.


From our perspective, there are distances up to 150m, but the largest distances are not so common, so that is why I proposed the 100m.


Whatever is not covered by the 0-100m, should be covered by the next PMD (most likely SMF based). That will reduce the cost by giving it additional volume, while the short distance on the MMF will drive the cost by increasing the yield.


Regards,


Peter



Petar Pepeljugoski
IBM Research
P.O.Box 218 (mail)
1101 Kitchawan Road, Rte. 134 (shipping)
Yorktown Heights, NY 10598

e-mail:
petarp@us.ibm.com
phone: (914)-945-3761
fax:        (914)-945-4134


Brad Booth <bbooth@QUAKETECH.COM >

08/22/2006 01:56 PM


Please respond to
Brad Booth <
bbooth@QUAKETECH.COM>


To
STDS-802-3-HSSG@listserv.ieee.org
cc
 
Subject
Re: [HSSG] Reach Objectives

 


   





Excuse my ignorance, but why?  When 10G started, this was an important debate.  If they had stuck with the number of PMDs in 802.3z, then there would have only been two PMDs.  The study group needs to define the markets that it wants to satisfy, and only then when the study group becomes a task force can the group decide on the actual number of PMDs required.

 

Thanks,

Brad




From: Drew Perkins [mailto:dperkins@infinera.com]
Sent:
Tuesday, August 22, 2006 10:45 AM
To:
Brad Booth;
STDS-802-3-HSSG@listserv.ieee.org
Subject:
RE: [HSSG] Reach Objectives


I have a simple proposal to save everyone a lot of time. Let's make the baseline plan to stick with exactly the same number of PMDs with exactly the same reaches as 10GbE. They have all been developed to fit someone's requirements. We should have a very high hurdle to eliminate any of these or add to them.

 
Drew

_____________________________

 
Drew Perkins

Chief Technology Officer

Infinera Corporation

1322 Bordeaux Drive

Sunnyvale, CA  94089

 
Phone:  408-572-5308

Cell:       408-666-1686

Fax:        408-904-4644

Email:    
dperkins@infinera.com
WWW :  
http://www.infinera.com
 
_____________________________

 

 



From:
Brad Booth [mailto:
bbooth@QUAKETECH.COM]
Sent:
Tuesday, August 22, 2006 10:19 AM
To:
STDS-802-3-HSSG@listserv.ieee.org
Subject:
Re: [HSSG] Reach Objectives

 

The 10km vs. 2km debate occurred in 802.3ae.  It was determined that the solution for 2km was virtually the same as that for 10km; therefore, the task force felt it would be better to select only one of those reaches and 10km was it.  If there is a desire for 2km and 10km reaches in HSSG, then in my humble opinion, those wishing to support will need to bring justification forward as to why the study group should consider possibly increasing the number of PHYs.

 

Cheers,

Brad

 

 



From:
Frank Chang [mailto:
ychang@VITESSE.COM]
Sent:
Tuesday, August 22, 2006 9:19 AM
To:
STDS-802-3-HSSG@listserv.ieee.org
Subject:
Re: [HSSG] Reach Objectives

John/Petar;

 

I agree datacenter for up to 100m. How about 10GigE LR disctance (10km) or campus 2km discussed earlier?

 

Frank

-----Original Message-----
From:
Petar Pepeljugoski [mailto:
petarp@US.IBM.COM]
Sent:
Tuesday, August 22, 2006 6:02 AM
To:
STDS-802-3-HSSG@listserv.ieee.org
Subject:
Re: [HSSG] Reach Objectives


John,


I believe that the Data Center distances should be 100m and 300m.

Petar Pepeljugoski
IBM Research
P.O.Box 218 (mail)
1101 Kitchawan Road, Rte. 134 (shipping)
Yorktown Heights, NY 10598

e-mail:
petarp@us.ibm.com
phone: (914)-945-3761
fax:        (914)-945-4134

John DAmbrosia <jdambrosia@FORCE10NETWORKS.COM >

08/22/2006 12:37 AM

 


Please respond to
John DAmbrosia <
jdambrosia@FORCE10NETWORKS.COM>

 


To
STDS-802-3-HSSG@listserv.ieee.org
cc
 
Subject
[HSSG] Reach Objectives


 

 


   





All,

We have had some conversation on the reflector regarding reach objectives.  Summarizing what has been discussed on the reflector I see the following –


Reach Objectives

Long-Haul   --> 100+ km

Metro       --> 10+ km

Data Center --> 50m & 300m


Data Center Reach Segregation

Intra-rack

Inter-rack

Horizontal runs

Vertical risers


Use this data to identify a single low-cost solution that would address a couple of the reach objectives


Other Areas

During the course of the CFI there were individuals who wanted Backplane Applications kept in for consideration, but I have not heard any further input in this area.  Are there still individuals who wish to propose Backplane as an objective?


John