[50762] in North American Network Operators' Group
Re: Do ATM-based Exchange Points make sense anymore?
daemon@ATHENA.MIT.EDU (Alex Rubenstein)
Thu Aug 8 12:37:55 2002
Date: Thu, 8 Aug 2002 12:37:30 -0400 (Eastern Daylight Time)
From: Alex Rubenstein <alex@nac.net>
To: "michael.dillon@radianz.com" <michael.dillon@radianz.com>
Cc: "nanog@merit.edu" <nanog@merit.edu>
In-Reply-To: <OF11688D95.3A1C38FD-ON80256C0F.0058C844-80256C0F.005966A6@radianz.com>
Errors-To: owner-nanog-outgoing@merit.edu
I have no first-hand experience SRP/DPT.
However, the following questions exist?
Does it scale well to 10's, or perhaps hundreds of nodes?
And, is there the possibility of one member hurting the entire ring?
On Thu, 8 Aug 2002, michael.dillon@radianz.com wrote:
>
> Given that your analysis could be used to make decisions about exchange
> point architecture, I wonder if you would consider adding RPR IEEE 802.17
> into the mix?
> While I don't know of anyone using RPR in an exchange point, there
> certainly are products on the market and not just Cisco's SRP/DPT. Within
> Ebone we were using SRP/DPT rings quite successfully in our PoP
> architecture, i.e. it does work in the real world. I know that the
> marketing side of RPR is promoting it as a metro area solution, but it
> works just as well in a PoP. Given that you are comparing a PoP oriented
> technology (Ethernet) with a metro area technology (ATM), I think it makes
> sense to take a serious look at a technology that covers both.
>
> --Michael Dillon
>
-- Alex Rubenstein, AR97, K2AHR, alex@nac.net, latency, Al Reuben --
-- Net Access Corporation, 800-NET-ME-36, http://www.nac.net --
