clause 38 methodology. Why deviate from this procedue? For example, Table 38.2 is as follows:
| Thread Links | Date Links | ||||
|---|---|---|---|---|---|
| Thread Prev | Thread Next | Thread Index | Date Prev | Date Next | Date Index |
My suggestion for HSSG distance objectives for fiber support is to partition
support requirements into "cells" per 1000BASE-X
clause 38 methodology. Why deviate from this procedue? For example,
Table 38.2 is as follows:
(Hopefully, most of you can see the above table. This is the first time
I have cut and pasted an Adobe Acrobat graphic into an email.
What a neat feature! Sorry if this is old news.)
The center column table cells indicate the fiber performance in terms of a bandwidth*distance product and is applicable to all PHY signaling schemes. The minimum range is either not necessary, or must be specified as a cell for each PHY signaling variant. My choice would be to eliminate it as redundant and confusing information. The only modification I can think of is the blending of results from the work of the TIA FO2.2 regarding laser lunch conditions which will likely affect the fiber performance parameter.
This methodology eliminates the need to interpret ill-defined terms
such as:
- the installed cable plant
- the equipment room
- a building/horizontal run
- multiple floors in a building/vertical/riser
- a campus
- etc.
Please consider the following wording for a distance motion:
That distance be specified in the form of a table with each row containing a fiber type and its associated fiber performance.
Best Regards,
Rich
--
"Cornejo, Edward (Edward)" wrote:
Brian, et al,-------------------------------------------------------------Thanks for your diligence in pressing this issue. Sorry, but I was not part
of this reflector until very recently. So here is some feedback on your
comments, and some additional comments.I don't see the value of having a general objective that states support for
traditional LAN, MAN, and WAN because it doesn't say anything if you don't
include distance and media type. Therefore, I favor including the
application, distance, and media type in the objective.I have a problem with at least 300m on MMF, and supporting the installed
base being added in the objective. Here is why:* I believe mentioning the installed base in the objective could be
too restrictive and possibly delay the release of the 10G standard. For
example, the serial proposal over new MMF using VCSEL technology could be
availble first, but would not satisfy the objective.
* We've had surprises with GbE regarding the installed base and it
would not be prudent to include it in the objective. Possible surprises for
10G can come from the use of coherent sources like DFBs over MMF.
* Lastly, having a minimum distance of 300m may exclude lower cost
technologies for shorter distance applications.My suggestion, which would not exclude the installed base, is as follows:
For Enterprise Applications
* At least 100m over multimode fiber for equipment room applications
* At least 200m over multimode fiber for riser applications
* We could always increase this number to 300m if everyone
feels confident their solution could support this. This just allows some
breathing room as proposals are fine tuned.
* At least 2km over single mode for the campus application
* The reason I want to throttle this back is because you could
be excluding lower cost serial solutions for this application. Also, based
on Chris Dominico's installed based study, he presented data that 2km's
satisfied a majority of the campus link distance requirements.
* Also, 2km is a distance that is defined for short reach in
other applications
* The reasons for this are clear; it is all about
allowing lower cost solutions to the customerFor MAN Applications
* At least 15km over single mode fiber for the intermediate
applications
* At least 40km over single mode fiber for the long reach applications
* At least 80km over single mode fiber for the extended reach
applicationsSince I have not been part of your earlier discussions, could someone
forward the details for the call this coming Monday.Thanks,
Ed C.-Lucent> ----------
> From:
> BRIAN_LEMOFF@xxxxxxxxxxxxxxxxxxxxxxxxxx[SMTP:BRIAN_LEMOFF@HP-PaloAlto-om16
> .om.hp.com]
> Sent: Friday, June 25, 1999 12:31 PM
> Cc: stds-802-3-hssg-distance@xxxxxxxx
> Subject: Distance Objectives !!!!!
>
>
> Three days ago, I suggested a distance objective on the HSSG distance
>
> reflector (the first suggestion, I believe, since Idaho). The purpose
>
> of this reflector is to provide a forum for the ad hoc distance
> committee (all of us receiving this e-mail) to reach a consensus
> prior
> to the Montreal meeting. There is a conference call Monday to reach
> "final" agreement. So far, there has been no response to my
> suggested
> objectives. Does this imply that we are all in agreement with them?
> If not, then perhaps someone who differs with me could offer a
> suggested alternative. Otherwise, we will simply pick up in Montreal
>
> where we left off in Idaho.
>
> -Brian Lemoff
> HP Labs
>
>
> ______________________________ Reply Separator
> _________________________________
> Subject: Re: Discussion on the HSSG Distance reflector
> Author: BRIAN-LEMOFF-at-om (BRIAN_LEMOFF@xxxxxxxxxxxxxxxxxxxxxxxxxx) at
> HP-PaloAlto,mimegw2
> Date: 6/22/99 11:47 AM
>
>
>
> As I understood Jonathan's kick-off message, he was proposing that we
>
> have two separate objectives, the first being phrased in general
> terms
> such as the one he proposed, i.e. support for traditional LAN,
> support
> for traditional MAN, and support for WAN access. The second, and
> more
> controversial objective will name the distance and fiber-type
> objectives. This is probably the more important of the two, and it
> has not yet been discussed on this reflector!!
>
> We might already have a consensus on the first objective (although
> there is still debate over support for the WAN), but we are nowhere
> near consensus on the second objective. I would support something
> like the following:
>
> To define physical solutions to support distances of:
>
> At least 300 m on multimode fiber, including the installed base
>
> At least 6 km on single mode fiber
>
> At least 50km on single mode fiber
>
>
> I am not wedded to these numbers, but I do think that some mention
> of
> support for the installed base should be included in this objective.
>
>
> -Brian Lemoff
> HP Labs
>
>
> ______________________________ Reply Separator
> _________________________________
> Subject: Discussion on the HSSG Distance reflector
> Author: del-hanson-at-exch (del_hanson@xxxxxxxxxxxxxx) at
> HP-PaloAlto,exchgw2
> Date: 6/22/99 11:07 AM
>
>
> HSSG Distance Ad Hoc Colleagues,
>
> I am on vacation this week but am following my email. I have noticed that
> the dialog on the Distance reflector has died up. If this is because there
> is agreement of the restated objectives, that is great. However, by
> sending
> out an announcement of a conference call for Monday 6/28/99 at 8 AM PST
> and
> an updated summary of objectives, it was not my intent to close down
> discussions. If there are issues that need to be discussed prior to the
> call, we need to use this reflector to make them visible. Also, after the
> call we need to use the reflector to refine issues so we can go into the
> Plenary with consensus, if possible.
>
> Regards,
> Del Hanson
>
Rich,
Your suggestion sounds more like language that would be included in a
specification within a standard, and less like an objective, whose purpose
is to define the mission of and set the goals for the group. Once we have
evaluated all of the available PMDs and how they perform over all of the
available fibers, we will be in a position to make a table like the one you
have included here. The question we need to ask ourselves is "What should
we set as the minimum we are willing to settle for in terms of distance and
fiber type". I feel that excluding support for the installed base of MMF
from the objectives will allow us to settle for a standard that falls short
of what our customers may be expecting. I have no objection to adding
additional distances and fiber types into the objectives, as long as we are
confident that we will be able to support them. Here is an example that
includes some of Ed Cornejo's suggestions:
To define physical solutions to support distances of:
At least 100 m on multimode fiber
At least 250 m on multimode fiber, including the installed base
At least 2 km on single mode fiber
At least 10 km on single mode fiber
At least 50km on single mode fiber
I personally don't feel that all of these are needed in the objectives
(particularly since the WWDM transceiver that we have been working on
should support the first four cases in a cost-effective manner), however if
this makes the objective more palatable to 75% of the committee, I have no
problem with it at all. I do feel strongly that support for installed MMF
should be included in the objective.
-Brian Lemoff
HP Labs
I concurr with Ed Cornejo's reasoning of including the distance, and media
type in the objective. The application, I believe, is
unnecessary. I also strongly agree with Ed's objection to a distance
support of 300 meters on MMF... if the intention is to support the
installed base of fiber. This unnecessarily precludes several signaling
proposals, is much greater than the distance support
requirements at 1/10th the speed (Gigabit Ethernet) and does not consider major
improvements in MMF cable technology which
may lead to significantly lower cost 10 GbE products. These three reasons are in
addition to the three that Ed offers in objection in
his note.
My suggestion for HSSG distance objectives for fiber support is to partition
support requirements into "cells" per 1000BASE-X
clause 38 methodology. Why deviate from this procedue? For example, Table 38.2
is as follows:
(Hopefully, most of you can see the above table. This is the first time I have
cut and pasted an Adobe Acrobat graphic into an email.
What a neat feature! Sorry if this is old news.)
The center column table cells indicate the fiber performance in terms of a
bandwidth*distance product and is applicable to all PHY signaling schemes. The
minimum range is either not necessary, or must be specified as a cell for each
PHY signaling variant. My choice would be to eliminate it as redundant and
confusing information. The only modification I can think of is the blending of
results from the work of the TIA FO2.2 regarding laser lunch conditions which
will likely affect the fiber performance parameter.
This methodology eliminates the need to interpret ill-defined terms such as:
- the installed cable plant
- the equipment room
- a building/horizontal run
- multiple floors in a building/vertical/riser
- a campus
- etc.
Please consider the following wording for a distance motion:
That distance be specified in the form of a table with each row containing a
fiber type and its associated fiber performance.
Best Regards,
Rich
--
"Cornejo, Edward (Edward)" wrote:
Brian, et al,
Thanks for your diligence in pressing this issue. Sorry, but I was not part
of this reflector until very recently. So here is some feedback on your
comments, and some additional comments.
I don't see the value of having a general objective that states support for
traditional LAN, MAN, and WAN because it doesn't say anything if you don't
include distance and media type. Therefore, I favor including the
application, distance, and media type in the objective.
I have a problem with at least 300m on MMF, and supporting the installed
base being added in the objective. Here is why:
* I believe mentioning the installed base in the objective could be
too restrictive and possibly delay the release of the 10G standard. For
example, the serial proposal over new MMF using VCSEL technology could be
availble first, but would not satisfy the objective.
* We've had surprises with GbE regarding the installed base and it
would not be prudent to include it in the objective. Possible surprises for
10G can come from the use of coherent sources like DFBs over MMF.
* Lastly, having a minimum distance of 300m may exclude lower cost
technologies for shorter distance applications.
My suggestion, which would not exclude the installed base, is as follows:
For Enterprise Applications
* At least 100m over multimode fiber for equipment room applications
* At least 200m over multimode fiber for riser applications
* We could always increase this number to 300m if everyone
feels confident their solution could support this. This just allows some
breathing room as proposals are fine tuned.
* At least 2km over single mode for the campus application
* The reason I want to throttle this back is because you could
be excluding lower cost serial solutions for this application. Also, based
on Chris Dominico's installed based study, he presented data that 2km's
satisfied a majority of the campus link distance requirements.
* Also, 2km is a distance that is defined for short reach in
other applications
* The reasons for this are clear; it is all about
allowing lower cost solutions to the customer
For MAN Applications
* At least 15km over single mode fiber for the intermediate
applications
* At least 40km over single mode fiber for the long reach applications
* At least 80km over single mode fiber for the extended reach
applications
Since I have not been part of your earlier discussions, could someone
forward the details for the call this coming Monday.
Thanks,
Ed C.-Lucent
> ----------
> From:
> BRIAN_LEMOFF@xxxxxxxxxxxxxxxxxxxxxxxxxx[SMTP:BRIAN_LEMOFF@HP-PaloAlto-om16
> .om.hp.com]
> Sent: Friday, June 25, 1999 12:31 PM
> Cc: stds-802-3-hssg-distance@xxxxxxxx
> Subject: Distance Objectives !!!!!
>
>
> Three days ago, I suggested a distance objective on the HSSG distance
>
> reflector (the first suggestion, I believe, since Idaho). The purpose
>
> of this reflector is to provide a forum for the ad hoc distance
> committee (all of us receiving this e-mail) to reach a consensus
> prior
> to the Montreal meeting. There is a conference call Monday to reach
> "final" agreement. So far, there has been no response to my
> suggested
> objectives. Does this imply that we are all in agreement with them?
> If not, then perhaps someone who differs with me could offer a
> suggested alternative. Otherwise, we will simply pick up in Montreal
>
> where we left off in Idaho.
>
> -Brian Lemoff
> HP Labs
>
>
> ______________________________ Reply Separator
> _________________________________
> Subject: Re: Discussion on the HSSG Distance reflector
> Author: BRIAN-LEMOFF-at-om (BRIAN_LEMOFF@xxxxxxxxxxxxxxxxxxxxxxxxxx) at
> HP-PaloAlto,mimegw2
> Date: 6/22/99 11:47 AM
>
>
>
> As I understood Jonathan's kick-off message, he was proposing that we
>
> have two separate objectives, the first being phrased in general
> terms
> such as the one he proposed, i.e. support for traditional LAN,
> support
> for traditional MAN, and support for WAN access. The second, and
> more
> controversial objective will name the distance and fiber-type
> objectives. This is probably the more important of the two, and it
> has not yet been discussed on this reflector!!
>
> We might already have a consensus on the first objective (although
> there is still debate over support for the WAN), but we are nowhere
> near consensus on the second objective. I would support something
> like the following:
>
> To define physical solutions to support distances of:
>
> At least 300 m on multimode fiber, including the installed base
>
> At least 6 km on single mode fiber
>
> At least 50km on single mode fiber
>
>
> I am not wedded to these numbers, but I do think that some mention
> of
> support for the installed base should be included in this objective.
>
>
> -Brian Lemoff
> HP Labs
>
>
> ______________________________ Reply Separator
> _________________________________
> Subject: Discussion on the HSSG Distance reflector
> Author: del-hanson-at-exch (del_hanson@xxxxxxxxxxxxxx) at
> HP-PaloAlto,exchgw2
> Date: 6/22/99 11:07 AM
>
>
> HSSG Distance Ad Hoc Colleagues,
>
> I am on vacation this week but am following my email. I have noticed that
> the dialog on the Distance reflector has died up. If this is because there
> is agreement of the restated objectives, that is great. However, by
> sending
> out an announcement of a conference call for Monday 6/28/99 at 8 AM PST
> and
> an updated summary of objectives, it was not my intent to close down
> discussions. If there are issues that need to be discussed prior to the
> call, we need to use this reflector to make them visible. Also, after the
> call we need to use the reflector to refine issues so we can go into the
> Plenary with consensus, if possible.
>
> Regards,
> Del Hanson
>
-------------------------------------------------------------
Richard Taborek Sr. Tel: 650 210 8800 x101 or 408 370 9233
Principal Architect Fax: 650 940 1898 or 408 374 3645
Transcendata, Inc. Email: rtaborek@xxxxxxxxxxxxxxxx
1029 Corporation Way http://www.transcendata.com
Palo Alto, CA 94303-4305 Alt email: rtaborek@xxxxxxxxxxxxx