Cell Relay Archive

Cell Relay Retreat>List Archive>month:1997-Feb> msg00280



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

SVC setup latency benchmarks?

  • From: John Sloan <jsloan@lucent.com>
  • Date: Thu, 27 Feb 1997 22:01:21 GMT

The topic is SVC setup latency. Can anyone suggest what factors
might effect the time it takes for an ATM switch to setup an SVC?
End to end latency depends individual switch latency and the number
of switches along the path, but what contributes to individual
switch SVC setup latency? Inter-arrival rates of setup requests?
Number of simultaneous SVCs already in service? Can anyone cite
any studies that have examined the curve of SVC setup latency
vs. (whatever metric is important)? Do folks have any sense whether
this curve is linear (doubtful IMHO), or exponential? How it may vary
from vendor to vendor depending on hardware/firmware architecture,
or from switch to switch with configuration? (Yeah, I know, it sounds
like I'm asking for the sun and the moon and a couple of gas giants,
but I'm trying to figure out what the conventional wisdom is -- which
chould probably be added to the FAQ -- and if any studies have been
done.)

Are SVCs just too too new (or non-existent) to have conventional
wisdom? !THANKS! in advance for any citations or insights.

-- 
John Sloan/DR32w56                                       +1 303 538 2746
Lucent Technologies/Bell Labs                        FAX +1 303 538 3155
11900 N. Pecos St.                              mailto:jsloan@lucent.com
Westminster CO 80234-2703 USA         http://info.dr.lucent.com/~jsloan/