Cell Relay Archive

Cell Relay Retreat>List Archive>month:1998-Jun> msg00130



[Date Prev][Date Next][Thread Prev][Thread Next]  
  [Date Index][Thread Index][Author Index][Subject Index]

[Q]: Division of work between CS and SAR sublayers in AAL?

  • From: Martin Steffen <ms@informatik.uni-kiel.de>
  • Date: 25 Jun 1998 15:56:35 +0200




Hi,

ploughing through different books about ATM, I found the division of work
between the convergence sublayer (CS) and the segmentation and reassembly
(SAR) sublayer quite confusing.

In several places it is mentioned that one of the task SAR has to perform,
is segementing longer messages into cells of the appropriate size (44 to 48
octets). This seems kind of natural considering the name ``SAR''.


On the other hand it is stated explicitely several times and explained in
detail (sometimes in the same books) that the convergence sublayer breaks
the incoming bitstream or messages coming from the higher layers into cells
and that it is those cells that are handed over from the CS down to the
SAR, and that the SAR sublayer only adds SAR-PDU headers/trailers to the
cells. This leaves then the question why the SAR layer is called
``segmentation and reassembly'' if this task is already performed by the
CS.

Does anybody know what is the right point of view there?

Thanks
	Martin