Cell Relay Archive

Cell Relay Retreat>List Archive>month:1999-Aug> msg00020



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

Re: AAL2 Header Processing

  • From: "SGRock" <auser@somedomain.com>
  • Date: Wed, 4 Aug 1999 18:02:19 GMT
  • Organization: NCCOSC RDT&E Division, San Diego, CA


I'm not sure about the AAL2 part of it, but depending on the physical medium
you're using, ATM cells can span frames (according to the UNI 3.1 spec).
For example, an ATM cell can span two STS-3 frames if the entire ATM cell
won't fit in the last row of the STS-3 payload.

My educated guess about your AAL2 question is that spanning is indeed
allowed, seeing as how the AAL PDU is chopped up to fit in the 48 byte ATM
payload.  If anyone can refute this assertion, please educate me.
SGRock

Pat Cauduro <patcc@earthlink.net> wrote in message
news:7nlci2$3a3$1@holly.prod.itd.earthlink.net...
> I have a fairly detailed question about implementing AAL2.
>
> Does anyone out there know if people pack cells such that the 3-byte
packet
> header actually spans ATM cells?  I see nothing in the standards that
> disallows this, but it would seem to complicate silicon implementations.
>
> Apologies if this is not a great forum for such a detailed question - let
me
> know if there's a better one!
>
> Thanks in advance -
>
>
> pat
>
> ____________________________________________
>
> Pat Cauduro
> patcc@earthlink.net
>
>
>
>