On Sun, 6 May 2001, Robert Davies wrote:
>
> ATM gets round this by mandating a pathetically small fixed cell size (I
> suppose you know how they arrived at it?) :)
ISTR it was a halfway house - arithmetic mean - between two opposing
camps, one of whom wanted larger (for data) and one smaller (for voice).
The basic premise was that data packets tend to take the same path and be
predominantly 'bursty' [0]; whereas voice (at the time since packet/cell
switching was in its' infancy) needed much smaller packets/cells to allow
for the more streamy nature of the information carried.
Also, the 53 bytes of an ATM cell carries (I can't remember how many!)
bytes of 'fraeme overhead' equivalent to the MAC header of an ethernet
packet, so the overall size for data far, far outweighs in percentage
terms the amount of user data inside it - and if you're doing VLANs over
ATM you lose even more of that space to framing info.
[0] this was, of course, in the days before anyone had brought streaming
audio/video to the end-user ;-)
I'm sure someone will correct me here, but that was what I understood
before we started ripping the ATM network out of Loughborough Uni a couple
of years ago - they're now entirely Gigabit Ethernet, in places from
server to desktop...
G
--------------------------------------------------------------------
http://www.lug.org.uk http://www.linuxportal.co.uk
http://www.linuxjob.co.uk http://www.linuxshop.co.uk
--------------------------------------------------------------------
This archive was generated by hypermail 2.1.3 : Thu 22 Nov 2001 - 13:12:00 GMT