Service provider T1/PPP question

Naslund, Steve SNaslund at medline.com
Fri Jun 28 19:56:57 UTC 2013


-----Original Message-----
From: Ricky Beam [mailto:jfbeam at gmail.com] 
Sent: Friday, June 28, 2013 2:45 PM
To: NANOG list; Mike
Subject: Re: Service provider T1/PPP question

On Fri, 28 Jun 2013 00:07:45 -0400, Mike <mike-nanog at tiedyenetworks.com>
wrote:
>> I am wanting to offer a broadband over T1 service and have the ...

>s/broadband/internet/

>A T1 is miles away from "broadband" these days.

>Having done this with Cisco gear (*years* ago), you want to avoid MLPPP whenever possible.  We did CEF per-packet when the CPE end was also Cisco; it worked perfectly with none of the restrictions or bugs.

>--Ricky

I think this post seems like a flashback.  I would not consider a T-1 to really be broadband anymore and it is pretty much limited to a business environment the way tariffs work.  As far as MLPPP, it seems to be pretty stable now where you need multiple bonded T-1s.  We have a few sites running MLPPP with Sprint on Juniper and Cisco gear and have not had an issue with it.  It is definitely not my preference for business connectivity anymore.  We tend to look for Ethernet service which is way cheaper per mb than T-1 and requires less expensive terminal equipment in most cases.  T-1s are the business solution where you need dedicated MPLS connectivity and fiber transport is not available.  DSL or Internet VPN are OK but somewhat less stable for business class private network solutions.  If it is internet connectivity they want you will get beaten up by the cable companies that can outrun and outprice you across the board.  You will also have a heck of a time competing with incumbent and competitive telecoms in T-1s that have central offices or collocations in central offices.  The economics just don't work if you don't have direct access to the cable plant.  Maybe up until the telecom act but not now.  How do you intend to get those T-1s back to you or are you a CLEC?

Steven Naslund





More information about the NANOG mailing list