Pointer to ATM performance work?
I seem to remember discussion at one of the meetings of some ATM performance work that looked at the benefits of discarding whole TCP packets when a cell was dropped. Can someone give me a pointer to this? --Richard
On Tue, 9 May 1995, Richard Colella wrote:
I seem to remember discussion at one of the meetings of some ATM performance work that looked at the benefits of discarding whole TCP packets when a cell was dropped. Can someone give me a pointer to this?
i think you're referring to Partial Packet Discard and Early Packet Discard. this is discussed in `Dynamics of TCP Traffic over ATM Networks' by Allyn Romanow and Sally Floyd which was published in SIGCOMM '94 in august. you can find this in <ftp://playground.sun.com/pub/tcp_atm> you also might want to check out the work by Floyd and Van Jacobson on Random Early Detection (RED) (volume 1, #4 of IEEE/ACM Transaction on Networking, august '93) - this one is available in <ftp://ftp.ee.lbl.gov/papers/early*> dsc
In message <9505091543.AA26046@snad.ncsl.nist.gov>, Richard Colella writes:
I seem to remember discussion at one of the meetings of some ATM performance work that looked at the benefits of discarding whole TCP packets when a cell was dropped. Can someone give me a pointer to this?
--Richard
Sally Floyd was coauthor of a paper with Allyn Rommanow that just got published in JSAC May 1995, "Dynamics of TCP Traffic over ATM". Since Sally coauthored, I usually look at ftp.ee.lbl.gov and get vectored to a sun site somewhere. My copy of README.papers is a old, but try playground.sun.com:pub/tcp_atm/tcpatm_dyn.0?.ps.Z where the ? is replaced with a document version. Curtis
participants (3)
-
Curtis Villamizar -
David Comay -
Richard Colella