[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: UP1500 Openssl speed




On Tue, 6 May 2003, Matt Simis wrote:

> 1x Athlon 2666MHz  (Barton, 512KB on chip L2)
>   sign     verify     sign/s      verify/s
> rsa  512 bits   0.0005s   0.0000s   2197.5  23393.1
> rsa 1024 bits   0.0021s   0.0001s    474.2   8645.8
> rsa 2048 bits   0.0123s   0.0004s     81.1   2728.6
> rsa 4096 bits   0.0811s   0.0013s     12.3    790.8
> RH9 supplied OpenSSL

...

> Not to bore everyone with x86 scores, but heres my other system, a P4 3.5GHz
> with HT enabled:
> 
>                   sign    verify    sign/s verify/s
> rsa  512 bits   0.0007s   0.0001s   1481.4  15356.9
> rsa 1024 bits   0.0033s   0.0002s    299.5   5359.6
> rsa 2048 bits   0.0204s   0.0006s     49.1   1613.2
> rsa 4096 bits   0.1379s   0.0022s      7.3    459.7
> Tests done on latest *Win32* version of OpenSSL, WindowsXP.
> 
> Quite appallingly bad isnt it? :D

It would be fair to see the test result on the same
P4 without HT enabled, before drawing any conclusions, as HT just divides
the cpu up into two virtual processors. But the openssl test does not seem
to be multithreaded.

Juhan Ernits



Reply to: