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

Re: what's next ?


I strongly disagree with your assessment of the outcome of the La Jolla meeting.
It is clear that the P802.3ae Task Force achieved virtually all of its objective
except for the specific identification of one or more out of three possible PMDs
to address multimode fiber optic cable objectives. All three of these PMDs
received significantly over a majority of the Task Force vote but was short of
the 75% required for passage.

I also disagree that LAN oriented proposals were "all voted down". I do not
understand how you can make this statement. All LAN and WAN objectives are
covered under the logic portion of the track and 100% of the objectives for both
LAN and WAN were voted in by the Task Force with virtually unanimous support.
The PMDs apply equally to the LAN and WAN. Ethernet in the WAN clearly requires
multimode fiber optic cable support. Do you disagree with this statement?

I disagree that there is a need for a separate PAR and its associated division
of effort at this time. I believe that the required compromise to bridge the
small gap between the current majority support and required 75% support will be
achieved in a reasonable period of time. In the interim, I am confident that
logic and PMD supporters of the only three candidates to satisfy the two
P802.3ae multimode cable plant objectives will not skip a beat in continuing to
optimize those PMD solutions to market. 

To conclude, I'd like to correct the PMDs referenced in your note to those
contending to meet P802.3ae multimode cable plant objectives and with
significantly more that majority support in the P802.3ae task force:

1. 850 nm Serial @ 10.3125 LAN/9.953 WAN line rate. PMD per kolesar_1_0700
2. 850 nm WDM (4 channel) @ 3.125 LAN/2.488 WAN line rate/channel. PMD per
3. 1310 nm WDM (4 channel) @ 3.125 LAN/2.488 WAN line rate/channel. PMD per

Note that the 850 nm Serial PMD leverages the same exact logic as voted in for
the accepted 1310 and 1550 nm Serial PMDs rather than reinventing the wheel with
an unsupported and much more difficult to achieve 12.5 Gbps data rate as you

To conclude, please note that specifications of two of three of the above PMDs
are currently being developed by ANSI accredited standards committee NCITS T11
in its T11.2 Optical Working Group. The remaining one will be proposed for
specification at the early August meeting of this committee. It has been
formally stated in the P802.3ae Task Force that these T11.2 specifications will
be "portable" to P802.3ae as input to that process.

A separate PAR is NOT needed. We won't skip a beat! 
Best Regards,
"Kardontchik.Jaime" wrote:
> Hello 10Giga'ers,
> The results of the vote held in La Jolla led to the ridiculous result
> that an
> 802.3 Task Force has passed the MAN- and WAN-oriented proposals
> and voted down all the LAN-oriented proposals. It is clear that the Task
> Force failed to reach its objectives. However, this failure could be
> also a
> great opportunity for everyone.
> Perhaps the right thing to do is to go for an amicable divorce and
> propose
> two new PARs and go for two separate Task Forces.
> One TF will have  MAN- and WAN-oriented objectives  using single-mode
> fiber and 1,300  and 1,550 nm lasers. These are the two proposals that
> were
> approved in La Jolla. Being freed from the "burden" of the multiplicity
> of
> LAN-oriented proposals (as some of its proponents declared) this TF
> could
> proceed unimpeded to reach all its desired objectives.
> The other TF will be LAN-oriented and will also proceed very rapidly to
> standardize the following 10.00000 Gbps proposals:
>     1) 8b/10b coding using 4-WDM and 850 nm lasers on multimode fiber,
> with 3.125 Gbaud symbol rate in the fiber;
>     2) 8b/10b coding using 4-WDM and 1,300 nm lasers on multimode and
> single-mode fiber, with 3.125 Gbaud symbol rate in the fiber; and
>     3) 8b/10b coding serially  with 12.5 Gbaud symbol rate in the fiber.
> The 4-months delay due to the need to get an approved PAR will not
> impact
> the final target schedule  of the original 802.3ae. On the contrary, one
> would
> expect - at least for the LAN-oriented Task Force - to have a first
> written
> draft and the first complete multivendor working prototypes by the end
> of
> this year (at least for the 4-WDM versions). This will make the approval
> of
> the LAN-PAR unstopable.  And with working prototypes so early one would
> expect the final LAN-oriented Standard to be very robust and overwhelmly
> adopted by the market.
> Moving the serial  LAN to 12.5 Gbaud is a risky proposition from my
> part. However,  if the delay/price penalty is reasonable it would be
> worthy
> since then all the 10 Gbps LAN PHYs will have the same PCS, that will
> also
> be shared by Fiber Channel and the majority of the high-speed Copper
> backplane solutions.
> These three LAN-oriented  proposals will provide the most cost effective
> solutions for all the possible LAN environments, including both the
> in-building
> links and the longer campus links
> Jaime E. Kardontchik
> Micro Linear
> San Jose, CA 95131
Richard Taborek Sr.                 Phone: 408-845-6102       
Chief Technology Officer             Cell: 408-832-3957
nSerial Corporation                   Fax: 408-845-6114
2500-5 Augustine Dr.        mailto:rtaborek@xxxxxxxxxxx
Santa Clara, CA 95054