[97889] in North American Network Operators' Group
Re: peter lothberg's mother slashdotted
daemon@ATHENA.MIT.EDU (micky coughes)
Fri Jul 13 08:23:41 2007
Date: Fri, 13 Jul 2007 08:21:16 -0400
From: "micky coughes" <coughes@gmail.com>
To: "Robert Blayzor" <rblayzor@inoc.net>
Cc: nanog@nanog.org
In-Reply-To: <46975F0F.4030707@inoc.net>
Errors-To: owner-nanog@merit.edu
------=_Part_31524_28635658.1184329276318
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: 7bit
Content-Disposition: inline
On 7/13/07, Robert Blayzor <rblayzor@inoc.net> wrote:
>
>
> micky coughes wrote:
> > I can see that *everybody* is missing the point on Peter's exercise.
> > Clearly this is to show to the telcos of the world that you can upgrade
> > to a native IP infrastructure and absorb the existing transport into the
> > router with a minimal effort. There was a post here from someone that
> > was there that explained how simple it was. This is HUGE! This has
> > the potential to completely disrupt telco transport dinosaur groups
> > *and* reshape the future. Taking it to his mom's house is just a poke
> > in the telco eye, he is making fun of them. This then begs the question
> > why can they do it between their facilities? If one guy can do it to a
> > *house* it must not be that hard. However, telcos with transport groups
> > of 1000s can't pull this off, this little project states volumes.
>
>
> There may be some telco's out there that don't know these types of
> technologies are out there. But for many, they are quiet aware. The
> fact is 10G seems a lot more economically feasible right now. Maybe if
> 40G, OC768 WDM line cards didn't cost over a quarter million dollars
> each there would be more deployments of it. The rate and cost of
> facility upgrades are far surpassing the means to make the ROI models
> work.
>
> -Robert
>
Of course they know 40G exists, read the press releases. It is how it is
implemented is the intersting thing.
Instead of relpacing all of DWDM systems from 10->40G it looks like it was
done to *one* window. Go run the nummbers on the follwoing models:
1. Replace 10G system to 40G + assorted new mux gear for existing circuits +
40G router card + opex
2. Remove one transponder + stratalight + 40G router card + opex
3. Existing 10G system + silly iMux 4x10G + 40G router card + opex (zero
gain in system capacity)
4. Remove one transponder + 40G DWDM router card + opex (the opex on this
one is huge because you start to elimiated a large number of bodies to
maintain #1, #2 or #3)
If you work for a telco go to your transport group and suggest #4 and listen
to their bell shaped heads ring.
------=_Part_31524_28635658.1184329276318
Content-Type: text/html; charset=ISO-8859-1
Content-Transfer-Encoding: 7bit
Content-Disposition: inline
<br><br>
<div><span class="gmail_quote">On 7/13/07, <b class="gmail_sendername">Robert Blayzor</b> <<a href="mailto:rblayzor@inoc.net">rblayzor@inoc.net</a>> wrote:</span>
<blockquote class="gmail_quote" style="PADDING-LEFT: 1ex; MARGIN: 0px 0px 0px 0.8ex; BORDER-LEFT: #ccc 1px solid"><br>micky coughes wrote:<br>> I can see that *everybody* is missing the point on Peter's exercise.<br>
> Clearly this is to show to the telcos of the world that you can upgrade<br>> to a native IP infrastructure and absorb the existing transport into the<br>> router with a minimal effort. There was a post here from someone that
<br>> was there that explained how simple it was. This is HUGE! This has<br>> the potential to completely disrupt telco transport dinosaur groups<br>> *and* reshape the future. Taking it to his mom's house is just a poke
<br>> in the telco eye, he is making fun of them. This then begs the question<br>> why can they do it between their facilities? If one guy can do it to a<br>> *house* it must not be that hard. However, telcos with transport groups
<br>> of 1000s can't pull this off, this little project states volumes.<br><br><br>There may be some telco's out there that don't know these types of<br>technologies are out there. But for many, they are quiet aware. The
<br>fact is 10G seems a lot more economically feasible right now. Maybe if<br>40G, OC768 WDM line cards didn't cost over a quarter million dollars<br>each there would be more deployments of it. The rate and cost of<br>
facility upgrades are far surpassing the means to make the ROI models work.<br><br>-Robert<br></blockquote></div>
<div><br>Of course they know 40G exists, read the press releases. It is how it is implemented is the intersting thing.</div>
<div> </div>
<div>Instead of relpacing all of DWDM systems from 10->40G it looks like it was done to *one* window. Go run the nummbers on the follwoing models:</div>
<div> </div>
<div>1. Replace 10G system to 40G + assorted new mux gear for existing circuits + 40G router card + opex</div>
<div>2. Remove one transponder + stratalight + 40G router card + opex</div>
<div>3. Existing 10G system + silly iMux 4x10G + 40G router card + opex (zero gain in system capacity)</div>
<div>4. Remove one transponder + 40G DWDM router card + opex (the opex on this one is huge because you start to elimiated a large number of bodies to maintain #1, #2 or #3)</div>
<div> </div>
<div>If you work for a telco go to your transport group and suggest #4 and listen to their bell shaped heads ring.</div>
<div> </div>
------=_Part_31524_28635658.1184329276318--
