Re: Long distance links
HSSG objectives only specify a MMF distance of 300M. How to you intend to meet the 500M
distance over MMF when operating at an OC-192 line rate? Do you have a 500M MMF WAN PHY
proposal available now or to present in York or Kauai?
Note the Lucent has presented a brand new fiber with a bandwidth*distance of 2200/500
MHz-km at 850/1300 nm with “laser” launch. This is one of the best fibers I know of.
It'll get you out to about 440M at an OC-192 line rate with VCSEL. Note also that this
is NEW not INSTALLED fiber.
In my estimation, if you want an inexpensive 500M MMF interface which connects a
customers LAN with DWDM equipment, you need to reduce the line rate with a MAS, WWDM,
or Parallel Optics approach.
Roy Bynum wrote:
> There are two different WAN specific implementation architectures. One of these is
> to use one of the SMF PHYs. The other uses a MMF PHY. When going into transport
> DWDM equipment at the customer or ISP POP site, the service providers would like to
> save money by using inexpensive interfaces also. The 500M MMF PHY is also a WAN
> Thank you,
> Roy Bynum
> MCI WorldCom
> Rich Taborek wrote:
> > Paul Bottorff wrote:
> > (text deleted)
> > > Provided people built networks to this configuration, then it works just
> > > fine.
> > > The IEEE has not yet decided to build 2 PHYs. I believe that the WAN PHY
> > > being talked about does not have a distinct identity from the LAN PHY.
> > > Because I don't have a good criteria for distinct identity I've found no
> > > reason to believe the committee should build 2 PHYs. My assumption is that
> > > any PHY developed may run on SMF and may be deployed in the wide area. This
> > > is what is currently happening with 1 GigE.
> > (text deleted)
> > Paul,
> > This is exactly what I'm concerned about. Your requirements to have Ethernet
> > operate at a MAC/PLS rate of 9.58464 Gbps, perform encoding with NRZ efficiency
> > (assumed to mean 0% overhead) and not use special symbols is unreasonable and
> > in conflict with virtually all 10 GbE PHY proposals presented to the HSSG to
> > date.
> > The best criteria for distinct identity is cost. For example, we can go through
> > all the components in a MAS PHY and all the components in your proposed WAN PHY
> > if you like. ...All right... I'll tell you: A prototype MAS PHY transceiver I
> > envision will use an SFF shell, a SFF connector, a MAS PHY CMOS chip including
> > laser driver, one laser, one photodiode/aIA/Post-amp and operate at 5 GBaud
> > (2.5 GHz). The result will be a significant difference in cost with a MAS PHY
> > coming out on top.
> > Another criteria is distance. A MAS PHY at 5 GBaud will support a MMF (all LANs
> > are MMF not SMF) distance of approximately 2X your proposed WAN PHY given the
> > same optics (which BTW need to support half the bandwidth of a WAN PHY and are,
> > therefore, less expensive.
> > Another criteria is ease of implementation through the use of integral clock
> > multiples in the most common 10 GbE equipment. That is, equipment which
> > supports multiple Ethernet data rates.
> > I could go on...
> > Many PHYs proposed for 10 GbE WILL NOT run over SMF. I'm probably splitting
> > hairs here, but VCSELs in general support only MMF. Many of the 10 GbE PHY
> > proposals support VCSELs. Are you excluding VCSELs from supporting 10 GbE?
> > I participated in the Gigabit Ethernet Standards process from day 0 till the
> > standard was published. There was never an objective set to support the WAN nor
> > was there any discussion or work done on a separate WAN PHY. As you and Roy
> > Bynum have pointed out, GbE is being successfully deployed in the WAN without
> > even considering this environment. I'm wondering right now if the HSSG should
> > do the same at 10 GbE?
> > Your proposal for a single PHY to meet all HSSG objectives including direct
> > support of the SONET WAN environment appears to be flawed from both a
> > cost/performance and simplicity perspective. 802.3 folks have drummed these
> > basic tenets into my head since I joined the group trying to sell them Fibre
> > Channel technology lock stock and barrel. I clearly changed my tune, proving my
> > flexibility. We can start negotiating any time now :-)
> > > Paul A. Bottorff, Director Switching Architecture
> > > Enterprise Solutions Technology Center
> > > Nortel Networks, Inc.
> > > 4401 Great America Parkway
> > > Santa Clara, CA 95052-8185
> > > Tel: 408 495 3365 Fax: 408 495 1299 ESN: 265 3365
> > > email: pbottorf@xxxxxxxxxxxxxxxxxx
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