[10583] in North American Network Operators' Group
Re: ATM vs. DS3
daemon@ATHENA.MIT.EDU (Ehud Gavron)
Thu Jul 10 01:10:02 1997
Date: Wed, 09 Jul 1997 19:52:23 -0700 (MST)
From: Ehud Gavron <GAVRON@ACES.COM>
In-reply-to: "Your message dated Wed, 09 Jul 1997 19:33:36 -0700 (PDT)"
<Pine.LNX.3.95q+pgcc+pgp.970709193103.12747A-100000@brap.connectnet.com>
To: Josh Beck <jbeck@connectnet.com>
Cc: nanog@merit.edu, GAVRON@ACES.COM
>Hello,
> I just thought of something. We are in the process of purchasing a
>4 Mb CIR from another backbone. Now, we have the choice of ATM or standard
>T3 delivery (over a DS3 either way). Now, if we get ATM, that 4 Mb CIR
>turns into:
>[ (53-5)/53 ] * 4 Mb/s = 48/53 * 4 Mb/s = 3.62 Mb/s
>Because an ATM cell is 53 bytes, but 5 bytes of that is header. Thus we
>pay for 4 Mb/s and get 3.62 Mb/s.
>Now, the question is, what is the overhead of the serial (ppp?) protocol
>running over the T3? Since it is point to point, and not addressed, it
>should be less, or is this not the case?
<24 bytes per MTU. If your MTU is 4000, your large frames
buy you 3976/4K, which conveniently multiplies by 4M to give you
3.9+M.
Also, don't forget that ATM switching introduces a latency that
a p2p connection doesn't.
If you're truly having a point-to-point application, and you have
a choice of a non-switched/routed protocol vs a p2p protocol...
um... go with b :)
E
>Any input would be appreciated!
>Josh Beck jbeck@connectnet.com
>------------------------------------------------------------------------
>CONNECTNet INS, Inc. Phone: (619)450-0254 Fax: (619)450-3216
>6370 Lusk Blvd., Suite F-208 San Diego, CA 92121
>------------------------------------------------------------------------
