home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.dcom.cell-relay
- Path: sparky!uunet!mcsun!dxcern!brian
- From: brian@dxcern.cern.ch (Brian Carpenter CERN-CN)
- Subject: Re: followup to reordering thread
- Message-ID: <1992Aug31.135521.14740@dxcern.cern.ch>
- Organization: CERN European Lab for Particle Physics
- References: <1992Aug30.164236.17945@sics.se>
- Date: Mon, 31 Aug 1992 13:55:21 GMT
- Lines: 29
-
- Another (very minor) disagreement with Craig. You should compute
- the AAL CRC on the fly as the cells arrive, not after the last cell.
- (I'm assuming you have hardware assist of course.) Otherwise, yes,
- switches cannot re-order ("Cell sequence integrity on a virtual channel
- connection is preserved by the ATM layer." - CCITT I.150, page 1.)
- and AAL5 is simple.
-
- -- Brian
-
- >> 2) If no re-ordering occurs, does that mean that all I have to do
- >> is collect the sequence of cells that have arrived on a VCI/VPI pair,
- >> perform some application (TCP,TP4 etc) dependent error control on the
- >> assembled packet, then just use the data (if re-transmission not needed).
- >
- >Yes. You receive cells and store them sequentially in a buffer. When
- >you have a complete PDU, you do the AAL CRC, and pass the packet up to
- >higher layers.
- >
- >> If the adaption layer is that simple under the no-reordering rule, then
- >> why not pay the price in additional switch sophistication and get the
- >> huge return in having a much simpler adaption layer?
- >>
- >> If re-ordering is then not allowed, then doesn't it make sense to have
- >> a end-of-message, begin-of-message bit in the cell header?
-
- >Yes. That's what AAL 5 is all about.
-
- >Craig
-
-