Re: UltraSPARC T2 - OpenSSL - PKCS11 ???
Thanks for your help everyone ! I've checked T5120 hardware specifications : 8 cores 8 threads/core so I used the multi 64 option : signverifysign/s verify/s rsa 512 bits 0.s 0.s 121283.7 202718.0 rsa 1024 bits 0.s 0.s 30643.1 115804.2 rsa 2048 bits 0.0002s 0.s 5282.8 44109.6 rsa 4096 bits 0.0421s 0.0009s 23.7 1071.3 It looks much better ! Does it look relevent to you ? About the multi 64 option : I tried multi 70 to be sure there will be always a thread waiting to be executed. I've noticed some improvement but not much ... Is this a good idea ? -- View this message in context: http://www.nabble.com/UltraSPARC-T2---OpenSSL---PKCS11-tp24952022p25002897.html Sent from the OpenSSL - User mailing list archive at Nabble.com. __ OpenSSL Project http://www.openssl.org User Support Mailing Listopenssl-users@openssl.org Automated List Manager majord...@openssl.org
UltraSPARC T2 - OpenSSL - PKCS11 ???
Hi, I'm currently working on a project involving a Sun T5120 and some crypto... (I'm also to to OpenSSL T2) T2 is supposed to be quite fast for crypto operations : openssl speed rsa signverifysign/s verify/s rsa 512 bits 0.0047s 0.0005s210.7 2072.0 rsa 1024 bits 0.0246s 0.0014s 40.7701.3 rsa 2048 bits 0.1554s 0.0048s 6.4206.3 rsa 4096 bits 1.0780s 0.0175s 0.9 57.0 openssl speed -engine pkcs11 rsa signverifysign/s verify/s rsa 512 bits 0.s 0.s 26514.1 31260.8 rsa 1024 bits 0.s 0.s 25112.3 30151.4 rsa 2048 bits 0.s 0.s 23563.2 29678.7 rsa 4096 bits 0.7007s 0.0184s 1.4 54.4 Compaired to the first speed test it looks pretty good ! To give me an idea how fast it is, I've run the same test on a desktop machine: signverifysign/s verify/s rsa 512 bits 0.000216s 0.15s 4637.7 67700.3 rsa 1024 bits 0.000835s 0.41s 1197.1 24520.1 rsa 2048 bits 0.004752s 0.000128s210.4 7809.1 rsa 4096 bits 0.030489s 0.000440s 32.8 2273.9 I guess something goes wrong ? I'm using the OpenSSL that was installed by default with Solaris 10 (it is supposed to be build with pkcs11 suport). Why is the T2 so slow (even when not using the pkcs11 engine) ? -- View this message in context: http://www.nabble.com/UltraSPARC-T2---OpenSSL---PKCS11-tp24952022p24952022.html Sent from the OpenSSL - User mailing list archive at Nabble.com. __ OpenSSL Project http://www.openssl.org User Support Mailing Listopenssl-users@openssl.org Automated List Manager majord...@openssl.org
Re: UltraSPARC T2 - OpenSSL - PKCS11 ???
On Thu, 13 Aug 2009, CH-Kami wrote: I'm currently working on a project involving a Sun T5120 and some crypto... (I'm also to to OpenSSL T2) hi, this is probably not the best alias since OpenSSL team has nothing to do with the PKCS#11 engine that is shipped with Solaris. T2 is supposed to be quite fast for crypto operations : openssl speed rsa signverifysign/s verify/s rsa 512 bits 0.0047s 0.0005s210.7 2072.0 rsa 1024 bits 0.0246s 0.0014s 40.7701.3 rsa 2048 bits 0.1554s 0.0048s 6.4206.3 rsa 4096 bits 1.0780s 0.0175s 0.9 57.0 openssl speed -engine pkcs11 rsa signverifysign/s verify/s rsa 512 bits 0.s 0.s 26514.1 31260.8 rsa 1024 bits 0.s 0.s 25112.3 30151.4 rsa 2048 bits 0.s 0.s 23563.2 29678.7 rsa 4096 bits 0.7007s 0.0184s 1.4 54.4 you must always specify -elapsed and run it on an idle computer if HW crypto is used. The time spent used to compute the numbers is in the user space (while you run it for 10 seconds) so you got wrong results (better results that you should have). Compaired to the first speed test it looks pretty good ! To give me an idea how fast it is, I've run the same test on a desktop machine: signverifysign/s verify/s rsa 512 bits 0.000216s 0.15s 4637.7 67700.3 rsa 1024 bits 0.000835s 0.41s 1197.1 24520.1 rsa 2048 bits 0.004752s 0.000128s210.4 7809.1 rsa 4096 bits 0.030489s 0.000440s 32.8 2273.9 I guess something goes wrong ? I don't think so. You desktop machine was 100% CPU bound during that test, unless you have more CPUs. T2 was idle, and utilizing 1 crypto core only (and not even fully). It has 8 crypto cores that can run in parallel. also, you may ignore RSA in 512 bits since that's not used nowadays. The Crypto Framework has some overhead so for operations that are short the overhead can be a significant part of the overall time. also note that T2 machines fit great in environments where things can be paralelized and crypto is needed; an SSL webserver, for example, but not for a small number of single-threaded tasks. cheers, J. -- Jan Pechanec __ OpenSSL Project http://www.openssl.org User Support Mailing Listopenssl-users@openssl.org Automated List Manager majord...@openssl.org
Re: [openssl-users] UltraSPARC T2 - OpenSSL - PKCS11 ???
Hi, Hodie Id. Aug. MMIX, CH-Kami scripsit: I'm currently working on a project involving a Sun T5120 and some crypto... (I'm also new to OpenSSL T2) T2 is supposed to be quite fast for crypto operations : openssl speed rsa signverifysign/s verify/s rsa 1024 bits 0.0246s 0.0014s 40.7701.3 openssl speed -engine pkcs11 rsa signverifysign/s verify/s rsa 1024 bits 0.s 0.s 25112.3 30151.4 Compaired to the first speed test it looks pretty good ! To give me an idea how fast it is, I've run the same test on a desktop machine: signverifysign/s verify/s rsa 1024 bits 0.000835s 0.41s 1197.1 24520.1 The T2 runs at a lower frequency, and you're testing only 1 core, and 1 thread. Do your tests using -multi 16 option, to saturate all your CPUs/cores/threads. On my laptop (Core2 duo 2GHz), this is what I get: signverifysign/s verify/s rsa 1024 bits 0.001340s 0.58s746.4 17313.2 Adjust the 16 to the number of CPUs in your T5200 (I tried with a T2000, equipped with 1 CPU, 4 cores, 4 threads/core). -- Erwann ABALEA erwann.aba...@keynectis.com - I can't be stupid, I completed third grade! __ OpenSSL Project http://www.openssl.org User Support Mailing Listopenssl-users@openssl.org Automated List Manager majord...@openssl.org
Re: UltraSPARC T2 - OpenSSL - PKCS11 ???
hi, your pkcs11 on the Sparc system is fast(!) its just the verification that seems a little b0rked/slow :-| alan __ OpenSSL Project http://www.openssl.org User Support Mailing Listopenssl-users@openssl.org Automated List Manager majord...@openssl.org