| |
 |
|
|
Science Forum Index » Cryptography Forum » OpenSSL Performance...
Page 1 of 1
|
| Author |
Message |
| Le Chaud Lapin... |
 Posted: Wed May 14, 2008 8:53 pm |
|
|
|
Guest
|
Hi All,
I did a quick sanity check on performance of my RSA implementation
with following parameters:
p 102639592829741105772054196573991675900716567808038066803341933521790711307779
q 106603488380168454820927220360012878679207958575989291522270608237193062808643
n = p * q [n has bit 511 set]
e = 17
I caculate [ciphertext = plaintext raised to e mod n] at just over
44,000 operations per second.
My code is un-optimized C++, with some assistive assembly, compiled
under Visual C++ 2005, running in Release mode on:
Microsoft® Windows Vista™ Ultimate
AMD Athlon(tm) 64 X2 Dual Core Processor 3600+, 1900 Mhz, 2 Core(s), 2
Logical Processor(s)
2.0 GB RAM
For a quick comparison I looked an OpenSSL performance table:
http://lists.freebsd.org/pipermail/freebsd-performance/2004-November/000956.html
The table shows 15545.4 OpenSSL verifications per second. As is the
case with so many crypto benchmarks, the table does not clearly
specifiy what exponent was used, etc. [drives me crazy when
crytographers do that:)], so I am assuming that verification was with
low-exponent = 17.
I will go over to OpenSSL site to get updated tables tomorrow, since
one above appears to be from 2004, but do these numbers seem right?
What exponent did they use?
-Le Chaud Lapin- |
|
|
| Back to top |
|
| Kristian Gjøsteen... |
| Posted: Wed May 14, 2008 10:27 pm |
|
|
|
Guest
|
Le Chaud Lapin <jaibuduvin at (no spam) gmail.com> wrote:
Quote: so I am assuming that verification was with
low-exponent = 17.
I don't know, but it may just as well be with 2^16+1.
--
Kristian Gjøsteen |
|
|
| Back to top |
|
| Le Chaud Lapin... |
| Posted: Tue May 20, 2008 6:42 pm |
|
|
|
Guest
|
On May 15, 3:27 am, Kristian Gjøsteen <kristiag+n... at (no spam) math.ntnu.no>
wrote:
Quote: Le Chaud Lapin <jaibudu... at (no spam) gmail.com> wrote:
so I am assuming that verification was with
low-exponent = 17.
I don't know, but it may just as well be with 2^16+1.
--
Kristian Gjøsteen
Thomas Pornin noted in a concurrent thread that, indeed, it is 65537,
which would explain the gross discrepancy between my performance and
theirs (mine is a lot slower).
-Le Chaud Lapin- |
|
|
| Back to top |
|
| |
|
Page 1 of 1
All times are GMT - 5 Hours
The time now is Fri Sep 05, 2008 1:15 am
|
|