Cell Relay Archive

["ronald h. davis" <ronaldd@lucent.com>]

Richard S. Tsumaki wrote:
> 
> · UBR specifies a PCR and CDVT, but in almost all implementations that's
> just set to the constraints of the physical interface for any VCC.
>

i'll go along with setting a pcr: it's an optional field but an actual
implementation would want to use a value for it.  setting cdvt, however,
doesn't make sense: the only reason for setting a cdvt limit is for
carrying traffic with real time constraints, like cbr and rt-vbr.

> · Barring any large buffers or packet discard schemes, UBR traffic from
> a high bit-rate source will suffer a high cell loss, if sent to a low
> bit-rate destination.  Any controls on the bit-rate are from the upper
> layer protocols/application at the ends of the connection.
> 

flow control is dependent upon the actions of the higher layer protocol.
depending upon how fast the response mechanism in that protocol is determines
high much cell loss is suffered.  i might add, though, that even if you
had an atm mechanism with flow control, such as abr, you would still have
the same problem of detection at the higher layer protocol since it's
flow control actions would be uncoordinated with those of atm.

-- 
  __  ______  __  / __/ | 	lucent technologies, naperville il, usa
_/ (_(_) / (_(_/_/_(_/  . 	ronald.h.davis@lucent.com
for there are many great deeds done in the small struggles of life.
life, misfortune, isolation, abandonment, poverty, are battlefields
which have their heros; obscure heroes, sometimes greater than the
illustrious heros.
				-victor hugo