Jean-Marc Desperrier wrote on Wed, 1 Sep 04, 3:45 AM:
> liupeng wrote:
> > I use webload 6.0 to do full SSL handshakes,SSL session can't reuse,and key
> > size=1024bit.
>
> Is "no reuse of SSL session" a realistic representation of how users
> will use your server ? Will you have many different user that do only
> one request and then their connexion is over ?
It depends on the application. A website which users connect to once then stay
connected for a long period will find that the reuse numbers more interesting
(applications: On-line banking/trading). If you have a system where the user connects
and grabs maybe one or two pages and that's it, the full handshake numbers are what
are most interesting (applications: shopping [only the checkout is SSL]).
> They are some test of SSL hardware accelerator cards around.
> They sometime show mixed results ; if you already have a lot of CPU
> power on your machine, some models of accelerators do not in fact
> actually accelerate it. It seems that the drivers are not always enough
> optimized so that the time spent communicating with the card can be
> longer than the time needed to do the calculation.
We found the most important point on getting good performance out of hardware
accelerators is getting good overlap. Most accelerators are really multi-processor
boxes. Often they have on par to slightly slower performance than modern host machines
for a single RSA operation. If you can overlap several operations at once, you start
to see single to double digit factors of performance improvement over unaccellerated
operations.
bob
> _______________________________________________
> mozilla-crypto mailing list
> [EMAIL PROTECTED]
> http://mail.mozilla.org/listinfo/mozilla-crypto
_______________________________________________
mozilla-crypto mailing list
[EMAIL PROTECTED]
http://mail.mozilla.org/listinfo/mozilla-crypto
