Q:Why router with ATM interface comes out earlier than pure SONET interface?

Mailman List Mirror (Read Only)
1

>One other fallacy here: ATM and POS do not provide the same effective
>bandwidth when used as an IP transport. Due to the high encapsulation
>overhead and the sizing overhead of fitting packets into cells, ATM has
>proven to be about 20% less efficient at carrying packets than SONET.
>While this is probably not an issue in a campus or within a building, the
>implications for long lines is enormous.
>

   Okay, so given all the great features that ATM is supposed to have
and the only thing that really sucks about it is the overhead due to the 53
byte cell size, the obvious question is why can't there be an ATM standard
with, say, 197 ( 4 times the current 48 byte payload) or even 389 ( 8
times 48 ) byte cells?
   Is there something magic about 53 or is the IP over ATM application
still so 'obscure' that there is no interest?

Increasing the cell size lowers the efficiency further.

53 is an ATM architectural constant. Change it, and it's no longer ATM.
Change it, and you're no longer interoperable.

Tony

2

> Okay, so given all the great features that ATM is supposed to have
> and the only thing that really sucks about it is the overhead
due to the 53
> byte cell size, the obvious question is why can't there be an
ATM standard
> with, say, 197 ( 4 times the current 48 byte payload) or even 389 ( 8
> times 48 ) byte cells?
> Is there something magic about 53 or is the IP over ATM application
> still so 'obscure' that there is no interest?

Increasing the cell size lowers the efficiency further.

53 is an ATM architectural constant. Change it, and it's no longer ATM.
Change it, and you're no longer interoperable.

Tony

Why not just make ATM variable cell size altogether?

Tongue planted firmly in cheek,
Chris

PS: Actually, overhead is not "the only thing that really sucks". Being
connection-oriented at the transport layer is another.

3

The old story was that the Telco guys wanted 32 byte payload and the data
guys wanted a 64 byte payload and the ITU split the difference. Go figure.

�� TORRENT NETWORKING TECHNOLOGIES CORP��
����������������������������������������
�� Next Generation Routing and Services��
�����������������������������������������
������� ����George Janosik���������������
��������� Sr Systems Engineer�������������
������ New Business Development���������
������������412.851.1103����������������
������ gjanosik@torrentnet.com����������
����������������������������������������
����� <http://www.torrentnet.com/>http://www.torrentnet.com
����������������������������������������

4

If the cell size was 1505 or larger would it still be lower efficiency
than 53 bytes? Sooner or later they'll probably try to come up with ATMng
won't they?

5

If the cell size was 1505 or larger would it still be lower efficiency
than 53 bytes? Sooner or later they'll probably try to come up with ATMng
won't they?

It is already here. Called IP. IP over SONET. With lots of IP QoS
mechanisms bolted on.

That is, if somebody can figure out a way to do private line (circuit
emulation) through an IP cloud. ATM will be an interim step, in my opinion.

Cheers,
Chris