Re: (SSIG) Taking the winning route
Ali,
I think there could be cases where it would make sense to install High BW multimode fiber if
there is a significant difference in the cost of PMDs as is the case of GbE with LX costing
approximately 4 to 5 times more than SX per port. Of course my assumption (shakey ground
now ..) is that the HBW fiber will cost nearly the same as current MMF prices (which still
costs more than single mode fiber). As for keeping track of the different installations, a
good documentation system goes a long way towards keeping it straight, but I know that is
often easier said than done. My point is there that if the price is right for the new
improved MMF, and there is a significant savings to the customer by using lower cost PMDs,
then why not provide them with the option to use it?
Respectfully,
Mike Bennett
Lawrence Berkeley Lab
ghiasi wrote:
> Hi Paul
>
> > From: "Kolesar, Paul F (Paul)" <pkolesar@xxxxxxxxxx>
> > To: "'ghiasi'" <Ali.Ghiasi@xxxxxxxxxxx>
> > Cc: stds-802-3-hssg@xxxxxxxx
> > Subject: RE: (SSIG) Taking the winning route
> > Date: Fri, 7 Apr 2000 12:22:04 -0400
> > MIME-Version: 1.0
> >
> > Ali,
> > I do not understand how you can be supportive of serial 850 nm
> > technology in one context and against it in another.
>
> I am supportive of 850 nm technology because I can operate a low cost VCSEL
> at 10 Gb/s with only few mA.
>
> You say "exploring 10
> > Gb serial with 850 nm VCSEL over 50/125 um at 100 m satisfies critical
> > market need at low cost." I agree. The same benefits apply to building
> > backbones up to 300 m in length.
>
> I am hesitant of pushing a new specialty fiber down the throat of my customers
> every few years. The standard should define the nominal distance based on existing
> multimode fiber, but individual manufactures may want to advertise extra reach with
> the new high bandwidth fiber.
>
> Why would I want to pull a new expensive high bandwidth multimode fiber, when it will be
> lower cost to use the existing single mode fiber.
>
> Finally how I am suppose to keep track of all these fibers in the field?
>
> Thanks,
>
> Ali Ghiasi
>
> Sun Microsystems
>
> >
> > Paul Kolesar
> >
> > ----------
> > From: ghiasi [SMTP:Ali.Ghiasi@xxxxxxxxxxx]
> > Sent: Friday, April 07, 2000 11:54 AM
> > To: nuss@xxxxxxxxxx; Kolesar, Paul F (Paul); avb@xxxxxxxxx
> > Cc: stds-802-3-hssg@xxxxxxxx; Ali.Ghiasi@xxxxxxxxxxx
> > Subject: Re: (SSIG) Taking the winning route
> >
> >
> > Hi Andy
> >
> > > Date: Fri, 7 Apr 2000 06:39:37 -0700 (PDT)
> > > From: Andreas Bechtolsheim <avb@xxxxxxxxx>
> > > To: nuss@xxxxxxxxxx, pkolesar@xxxxxxxxxx
> > > Subject: Re: (SSIG) Taking the winning route
> > > Cc: stds-802-3-hssg@xxxxxxxx
> > > X-Resent-To: Multiple Recipients
> > <stds-802-3-hssg@xxxxxxxxxxxxxxxxxx>
> > > X-Listname: stds-802-3-hssg
> > > X-Info: [Un]Subscribe requests to majordomo@xxxxxxxxxxxxxxxxxx
> > > X-Moderator-Address: stds-802-3-hssg-approval@xxxxxxxxxxxxxxxxxx
> > >
> > >
> > >
> > > 850nm VCSELs for 10 GE require customers to deploy new
> > high-bandwdith MMF.
> > > I would recommend that PMD solutions are based on installed base
> > MMF or SMF.
> >
> > I would agree that PMD solution should be based on existing MMF and
> > SMF. At the cost of
> > pulling new high bandwidth fiber, just pull SMF and operate at 1300
> > nm.
> >
> > >
> > > Specifying a PMD that implies a brand new MMF infrastructure is
> > not
> > > consistent with the goal of using existing technology wherever
> > possible.
> >
> > But exploring 10 Gb serial with 850 nm VCSEL over 50/125 um at 100 m
> > satisfies critical market
> > need at low cost.
> >
> > Thanks,
> >
> > Ali Ghiasi
> >
> > Sun Microsystems
> >
> >