Re: [802.3BA] Discussion on 40G for => 10 km SMF
There is nothing about any approach being considered for 40GE 10km SMF
that helps DWDM better then the other being considered.
We have already explained that both 10G 1310nm un-cooled DFB and 40G
1310nm cooled EML are not useful as DWDM sources. This is especially
true for 80km distance which is the starting point for useful DWDM
With respect to ICs, the usefulness depends on what type of 40G DWDM
approach is used.
If the approach is ODB or DPSK, then 40GE Serial ICs are most useful.
If the approach is DQPSK, then 2x20G (as proposed by Avago) ICs are most
If the approach is DP-QPSK or 4 sub-carriers then CWDM ICs are most
802.3ba is not the right place to debate the best approach to 40G DWDM.
There is no justification for adding any cost to 40GE solution to create
hypothetical synergy with DWDM, especially because there is none.
From: Mikael Abrahamsson [mailto:swmike@xxxxxxxxx]
Sent: Friday, August 08, 2008 2:09 PM
Subject: Re: [802.3BA] Discussion on 40G for => 10 km SMF
On Fri, 8 Aug 2008, Trowbridge, Stephen J (Steve) wrote:
> The colored interfaces you are describing are not even in the
> but they appear in the market.
Quite, and with the 4x10GE CWDM approach, these won't appear anytime
as (I would guess) it's easier to make a colored version of a serial
already existing interface, than to first develop a 40G serial and then
colored version of that.
> interface in the future, no doubt it will be developed. All people are
> saying is that based on cost and time to market, 40G serial is not the
> right solution to develop now for the first version of the standard.
> don't normally decide what direction to follow for a 2010 product
> on what we speculate someone might invent in 2013.
Well, it's always important to remember the con:s of ones decision, and
think colored version of existing standard available early is a door
shut if 4x10G CWDM is chosen.
Mikael Abrahamsson email: swmike@xxxxxxxxx