Andy Polyakov wrote:
http://netflow.internet2.edu/weekly/20070820/
Given this context I have to admit that I have effectively misused
"throughput" term in my posts. I should have written "amount of private
key operation requests per time unit" and speculate about its effect on
overall service time becoming unnecessary high at low request rate.
Oh I understood the context you meant. Basically I was just trying to
show, for at least this segment of the network, how much traffic was
being handled by which application. Obviously, as you point out, this is
an imperfect way to estimate things. It wasn't really meant to be
anything more than a data point as I'm not involved enough to weigh in
on one side or another on the issue.
Moving back to the original context the relevant data would be amount of
connections being established and *among them* amount of connections
triggering private key operations, not summed up TCP throughput or even
packet size distribution.
The data on the number of connections being made is likely being
captured as well. I could ask to make it available.
private key operations." In SSH context every connection established
triggers one, but connection rate is rather low as connections are
commonly held for relatively long times. In HTTPS context connection
rate can be high, but not every connection triggers private key
operation thanks to session context being reused for connections to same
server. A.
______________________________________________________________________
OpenSSL Project http://www.openssl.org
Development Mailing List [email protected]
Automated List Manager [EMAIL PROTECTED]
