Portál AbcLinuxu, 24. dubna 2024 23:20

Srovnání i5-2400@3.10GHz s Atomem C2750@2.40GHz

10.7.2015 01:00 | Přečteno: 1724× | poslední úprava: 10.7.2015 16:33

Dostal jsem se k serveru se základní deskou SuperMicro A1SA7-2750F osazenou osmijádrovým procesorem Intel(R) Atom(TM) CPU C2750 @ 2.40GHz a paměťmi DDR3 taktovanými na 1600MHz. A protože do budoucna uvažuji o tom, že ve svém domácím serveru vyměním základní desku za něco s podporou IPMI, řekl jsem si, že zkusím porovnat výkon se stávajícím HW, což je deska Gigabyte Z77MX-D3H osazená čtyřjádrovým procesorem Intel(R) Core(TM) i5-2400 CPU @ 3.10GHz bez hyperthreadingu a DDR3 RAM taktovanou na 1333 MHz.
Upozornění: v článku jsou jen surová data. Pokud chcete nějaké předžvýkané závěry, je pro vás zbytečné to otevírat...

Testování jsem - s ohledem na cílové využití - prováděl v KVM virtuálu, kterému jsem přidělil 4GB RAM a všechna dostupná jádra. Operačním systémem byl s výjimkou posledního testu aktuální Gentoo Linux; poslední test jsem spouštěl v Debianu Jessie v LXC kontejneru.

Jako první přišel na řadu sysbench, nejprve spuštěný v jednom vlákně a pak v osmi.

i5-2400:

# sysbench --test=cpu --cpu-max-prime=20000 run
sysbench 0.4.12:  multi-threaded system evaluation benchmark

Running the test with following options:
Number of threads: 1

Doing CPU performance benchmark

Threads started!
Done.

Maximum prime number checked in CPU test: 20000


Test execution summary:
 total time:                          26.0923s
 total number of events:              10000
 total time taken by event execution: 26.0891
 per-request statistics:
 min:                                  2.56ms
 avg:                                  2.61ms
 max:                                  3.74ms
 approx.  95 percentile:               2.66ms

Threads fairness:
 events (avg/stddev):           10000.0000/0.00
 execution time (avg/stddev):   26.0891/0.00

# sysbench --test=cpu --cpu-max-prime=20000 --num-threads=8 run
sysbench 0.4.12:  multi-threaded system evaluation benchmark

Running the test with following options:
Number of threads: 8

Doing CPU performance benchmark

Threads started!
Done.

Maximum prime number checked in CPU test: 20000


Test execution summary:
 total time:                          6.7482s
 total number of events:              10000
 total time taken by event execution: 53.9059
 per-request statistics:
 min:                                  2.64ms
 avg:                                  5.39ms
 max:                                 22.68ms
 approx.  95 percentile:              12.69ms

Threads fairness:
 events (avg/stddev):           1250.0000/3.43
 execution time (avg/stddev):   6.7382/0.01

Atom C2750:

# sysbench --test=cpu --cpu-max-prime=20000 run
sysbench 0.4.12:  multi-threaded system evaluation benchmark

Running the test with following options:
Number of threads: 1

Doing CPU performance benchmark

Threads started!
Done.

Maximum prime number checked in CPU test: 20000


Test execution summary:
 total time:                          42.6339s
 total number of events:              10000
 total time taken by event execution: 42.6243
 per-request statistics:
 min:                                  4.25ms
 avg:                                  4.26ms
 max:                                  8.47ms
 approx.  95 percentile:               4.27ms

Threads fairness:
 events (avg/stddev):           10000.0000/0.00
 execution time (avg/stddev):   42.6243/0.00

# sysbench --test=cpu --cpu-max-prime=20000 --num-threads=8 run
sysbench 0.4.12:  multi-threaded system evaluation benchmark

Running the test with following options:
Number of threads: 8

Doing CPU performance benchmark

Threads started!
Done.

Maximum prime number checked in CPU test: 20000


Test execution summary:
 total time:                          5.5612s
 total number of events:              10000
 total time taken by event execution: 44.4255
 per-request statistics:
 min:                                  4.26ms
 avg:                                  4.44ms
 max:                                 39.65ms
 approx.  95 percentile:               4.29ms

Threads fairness:
 events (avg/stddev):           1250.0000/37.79
 execution time (avg/stddev):   5.5532/0.00

Následovala klasická kompilace linuxového jádra, konfigurační soubor byl v obou případech stejný. Nejprve jsem kompiloval s parametry "-j5 -l5" (protože i5-2400 je čtyřjádro), posléze s "-j9 -l9" (protože Atom má osm jader).

Edit: Na návrh Michala Kubečka dole v diskusi jsem zkusil kompilovat na tmpfs (původně jsem si jen adresář se zdrojáky jádra natáhl do cache); na i5 byly rozdíly minimální, na Atomu se mi ovšem podařilo stáhnout původní skoro čtyřminutový náskok i5 téměř na nulu. Panu Kubečkovi díky za nakopnutí. Před samotným spuštěním genkernelu jsem ještě pokaždé spouštěl make mrproper a make clean, swap byl během kompilace vypnutý.

i5-2400:

# time genkernel --makeopts="-j5 -l5" --no-ramdisk-modules --lvm --no-mrproper --no-clean all
...
real    5m34.204s
user    18m42.920s
sys     1m31.889s
# time genkernel --makeopts="-j9 -l9" --no-ramdisk-modules --lvm --no-mrproper --no-clean all
...
real    5m38.718s
user    19m17.433s
sys     1m34.395s

Atom C2750:

# time genkernel --makeopts="-j5 -l5" --no-ramdisk-modules --lvm --no-mrproper --no-clean all
...
real    13m7.268s
user    58m35.740s
sys     5m41.435s
# time genkernel --makeopts="-j9 -l9" --no-ramdisk-modules --lvm --no-mrproper --no-clean all
...
real    5m39.494s
user    18m43.313s
sys     1m30.798s

Přidal jsem i jednovláknový benchmark pomocí OpenSSL:

i5-2400:

# openssl speed
Doing mdc2 for 3s on 16 size blocks: 2373881 mdc2's in 2.99s
Doing mdc2 for 3s on 64 size blocks: 626679 mdc2's in 3.00s
Doing mdc2 for 3s on 256 size blocks: 159622 mdc2's in 3.00s
Doing mdc2 for 3s on 1024 size blocks: 40125 mdc2's in 3.00s
Doing mdc2 for 3s on 8192 size blocks: 5023 mdc2's in 3.00s
Doing md4 for 3s on 16 size blocks: 16117042 md4's in 2.99s
Doing md4 for 3s on 64 size blocks: 11919173 md4's in 3.00s
Doing md4 for 3s on 256 size blocks: 6697254 md4's in 3.00s
Doing md4 for 3s on 1024 size blocks: 2424631 md4's in 3.00s
Doing md4 for 3s on 8192 size blocks: 347570 md4's in 3.00s
Doing md5 for 3s on 16 size blocks: 11767588 md5's in 3.00s
Doing md5 for 3s on 64 size blocks: 8301704 md5's in 3.00s
Doing md5 for 3s on 256 size blocks: 4454013 md5's in 2.99s
Doing md5 for 3s on 1024 size blocks: 1558348 md5's in 3.00s
Doing md5 for 3s on 8192 size blocks: 220365 md5's in 3.00s
Doing hmac(md5) for 3s on 16 size blocks: 9514608 hmac(md5)'s in 3.00s
Doing hmac(md5) for 3s on 64 size blocks: 7137773 hmac(md5)'s in 3.00s
Doing hmac(md5) for 3s on 256 size blocks: 4077195 hmac(md5)'s in 3.00s
Doing hmac(md5) for 3s on 1024 size blocks: 1507558 hmac(md5)'s in 3.00s
Doing hmac(md5) for 3s on 8192 size blocks: 219094 hmac(md5)'s in 3.00s
Doing sha1 for 3s on 16 size blocks: 13339301 sha1's in 2.99s
Doing sha1 for 3s on 64 size blocks: 9005669 sha1's in 3.00s
Doing sha1 for 3s on 256 size blocks: 4713340 sha1's in 3.00s
Doing sha1 for 3s on 1024 size blocks: 1626254 sha1's in 3.00s
Doing sha1 for 3s on 8192 size blocks: 229079 sha1's in 3.00s
Doing sha256 for 3s on 16 size blocks: 7732616 sha256's in 3.00s
Doing sha256 for 3s on 64 size blocks: 4107283 sha256's in 3.00s
Doing sha256 for 3s on 256 size blocks: 1723494 sha256's in 3.00s
Doing sha256 for 3s on 1024 size blocks: 517427 sha256's in 3.00s
Doing sha256 for 3s on 8192 size blocks: 68646 sha256's in 3.00s
Doing sha512 for 3s on 16 size blocks: 6049470 sha512's in 3.00s
Doing sha512 for 3s on 64 size blocks: 6059080 sha512's in 2.99s
Doing sha512 for 3s on 256 size blocks: 2204826 sha512's in 3.00s
Doing sha512 for 3s on 1024 size blocks: 761499 sha512's in 3.00s
Doing sha512 for 3s on 8192 size blocks: 106937 sha512's in 3.00s
Doing whirlpool for 3s on 16 size blocks: 4406999 whirlpool's in 3.00s
Doing whirlpool for 3s on 64 size blocks: 2349820 whirlpool's in 3.00s
Doing whirlpool for 3s on 256 size blocks: 971685 whirlpool's in 3.00s
Doing whirlpool for 3s on 1024 size blocks: 290419 whirlpool's in 3.00s
Doing whirlpool for 3s on 8192 size blocks: 38582 whirlpool's in 2.99s
Doing rmd160 for 3s on 16 size blocks: 7208626 rmd160's in 3.00s
Doing rmd160 for 3s on 64 size blocks: 4183391 rmd160's in 3.00s
Doing rmd160 for 3s on 256 size blocks: 1862861 rmd160's in 3.00s
Doing rmd160 for 3s on 1024 size blocks: 579032 rmd160's in 3.00s
Doing rmd160 for 3s on 8192 size blocks: 77938 rmd160's in 3.00s
Doing rc4 for 3s on 16 size blocks: 70074457 rc4's in 3.00s
Doing rc4 for 3s on 64 size blocks: 29468097 rc4's in 3.00s
Doing rc4 for 3s on 256 size blocks: 8584863 rc4's in 3.00s
Doing rc4 for 3s on 1024 size blocks: 2243867 rc4's in 3.00s
Doing rc4 for 3s on 8192 size blocks: 283013 rc4's in 3.00s
Doing des cbc for 3s on 16 size blocks: 11448337 des cbc's in 3.00s
Doing des cbc for 3s on 64 size blocks: 2959117 des cbc's in 2.99s
Doing des cbc for 3s on 256 size blocks: 744618 des cbc's in 3.00s
Doing des cbc for 3s on 1024 size blocks: 186183 des cbc's in 3.00s
Doing des cbc for 3s on 8192 size blocks: 23346 des cbc's in 3.00s
Doing des ede3 for 3s on 16 size blocks: 4437179 des ede3's in 3.00s
Doing des ede3 for 3s on 64 size blocks: 1125238 des ede3's in 3.00s
Doing des ede3 for 3s on 256 size blocks: 281642 des ede3's in 3.00s
Doing des ede3 for 3s on 1024 size blocks: 70784 des ede3's in 3.00s
Doing des ede3 for 3s on 8192 size blocks: 8817 des ede3's in 3.00s
Doing aes-128 cbc for 3s on 16 size blocks: 18805658 aes-128 cbc's in 3.00s
Doing aes-128 cbc for 3s on 64 size blocks: 5065686 aes-128 cbc's in 3.00s
Doing aes-128 cbc for 3s on 256 size blocks: 1286671 aes-128 cbc's in 3.00s
Doing aes-128 cbc for 3s on 1024 size blocks: 761409 aes-128 cbc's in 2.99s
Doing aes-128 cbc for 3s on 8192 size blocks: 96129 aes-128 cbc's in 3.00s
Doing aes-192 cbc for 3s on 16 size blocks: 15910118 aes-192 cbc's in 3.00s
Doing aes-192 cbc for 3s on 64 size blocks: 4143075 aes-192 cbc's in 3.00s
Doing aes-192 cbc for 3s on 256 size blocks: 1063572 aes-192 cbc's in 3.00s
Doing aes-192 cbc for 3s on 1024 size blocks: 640394 aes-192 cbc's in 3.00s
Doing aes-192 cbc for 3s on 8192 size blocks: 80986 aes-192 cbc's in 2.99s
Doing aes-256 cbc for 3s on 16 size blocks: 13734690 aes-256 cbc's in 3.00s
Doing aes-256 cbc for 3s on 64 size blocks: 3611446 aes-256 cbc's in 3.00s
Doing aes-256 cbc for 3s on 256 size blocks: 913283 aes-256 cbc's in 3.00s
Doing aes-256 cbc for 3s on 1024 size blocks: 554078 aes-256 cbc's in 3.00s
Doing aes-256 cbc for 3s on 8192 size blocks: 70116 aes-256 cbc's in 3.00s
Doing aes-128 ige for 3s on 16 size blocks: 19099022 aes-128 ige's in 3.00s
Doing aes-128 ige for 3s on 64 size blocks: 4876708 aes-128 ige's in 2.99s
Doing aes-128 ige for 3s on 256 size blocks: 1224969 aes-128 ige's in 3.00s
Doing aes-128 ige for 3s on 1024 size blocks: 307067 aes-128 ige's in 3.00s
Doing aes-128 ige for 3s on 8192 size blocks: 38462 aes-128 ige's in 3.00s
Doing aes-192 ige for 3s on 16 size blocks: 16072323 aes-192 ige's in 3.00s
Doing aes-192 ige for 3s on 64 size blocks: 4086543 aes-192 ige's in 3.00s
Doing aes-192 ige for 3s on 256 size blocks: 1024743 aes-192 ige's in 3.00s
Doing aes-192 ige for 3s on 1024 size blocks: 257446 aes-192 ige's in 3.00s
Doing aes-192 ige for 3s on 8192 size blocks: 32197 aes-192 ige's in 3.00s
Doing aes-256 ige for 3s on 16 size blocks: 13881367 aes-256 ige's in 2.99s
Doing aes-256 ige for 3s on 64 size blocks: 3526286 aes-256 ige's in 3.00s
Doing aes-256 ige for 3s on 256 size blocks: 884385 aes-256 ige's in 3.00s
Doing aes-256 ige for 3s on 1024 size blocks: 221056 aes-256 ige's in 3.00s
Doing aes-256 ige for 3s on 8192 size blocks: 27681 aes-256 ige's in 3.00s
Doing ghash for 3s on 16 size blocks: 163435996 ghash's in 3.00s
Doing ghash for 3s on 64 size blocks: 66049459 ghash's in 3.00s
Doing ghash for 3s on 256 size blocks: 18834625 ghash's in 3.00s
Doing ghash for 3s on 1024 size blocks: 4834211 ghash's in 3.00s
Doing ghash for 3s on 8192 size blocks: 616092 ghash's in 3.00s
Doing camellia-128 cbc for 3s on 16 size blocks: 17058902 camellia-128 cbc's in 3.00s
Doing camellia-128 cbc for 3s on 64 size blocks: 6473554 camellia-128 cbc's in 3.00s
Doing camellia-128 cbc for 3s on 256 size blocks: 1824265 camellia-128 cbc's in 2.99s
Doing camellia-128 cbc for 3s on 1024 size blocks: 472808 camellia-128 cbc's in 3.00s
Doing camellia-128 cbc for 3s on 8192 size blocks: 59820 camellia-128 cbc's in 3.00s
Doing camellia-192 cbc for 3s on 16 size blocks: 14832406 camellia-192 cbc's in 3.00s
Doing camellia-192 cbc for 3s on 64 size blocks: 5058798 camellia-192 cbc's in 3.00s
Doing camellia-192 cbc for 3s on 256 size blocks: 1382096 camellia-192 cbc's in 3.00s
Doing camellia-192 cbc for 3s on 1024 size blocks: 356415 camellia-192 cbc's in 3.00s
Doing camellia-192 cbc for 3s on 8192 size blocks: 44995 camellia-192 cbc's in 3.00s
Doing camellia-256 cbc for 3s on 16 size blocks: 14429891 camellia-256 cbc's in 3.00s
Doing camellia-256 cbc for 3s on 64 size blocks: 5061328 camellia-256 cbc's in 3.00s
Doing camellia-256 cbc for 3s on 256 size blocks: 1386924 camellia-256 cbc's in 3.00s
Doing camellia-256 cbc for 3s on 1024 size blocks: 356687 camellia-256 cbc's in 3.00s
Doing camellia-256 cbc for 3s on 8192 size blocks: 44948 camellia-256 cbc's in 3.00s
Doing idea cbc for 3s on 16 size blocks: 14327979 idea cbc's in 2.99s
Doing idea cbc for 3s on 64 size blocks: 3714940 idea cbc's in 3.00s
Doing idea cbc for 3s on 256 size blocks: 936766 idea cbc's in 3.00s
Doing idea cbc for 3s on 1024 size blocks: 234479 idea cbc's in 3.00s
Doing idea cbc for 3s on 8192 size blocks: 29431 idea cbc's in 3.00s
Doing seed cbc for 3s on 16 size blocks: 11929264 seed cbc's in 3.00s
Doing seed cbc for 3s on 64 size blocks: 3000230 seed cbc's in 3.00s
Doing seed cbc for 3s on 256 size blocks: 751131 seed cbc's in 3.00s
Doing seed cbc for 3s on 1024 size blocks: 187673 seed cbc's in 3.00s
Doing seed cbc for 3s on 8192 size blocks: 23471 seed cbc's in 3.00s
Doing rc2 cbc for 3s on 16 size blocks: 6957005 rc2 cbc's in 3.00s
Doing rc2 cbc for 3s on 64 size blocks: 1767965 rc2 cbc's in 3.00s
Doing rc2 cbc for 3s on 256 size blocks: 444792 rc2 cbc's in 2.99s
Doing rc2 cbc for 3s on 1024 size blocks: 111363 rc2 cbc's in 3.00s
Doing rc2 cbc for 3s on 8192 size blocks: 13919 rc2 cbc's in 3.00s
Doing rc5-32/12 cbc for 3s on 16 size blocks: 33417726 rc5-32/12 cbc's in 3.00s
Doing rc5-32/12 cbc for 3s on 64 size blocks: 9224343 rc5-32/12 cbc's in 3.00s
Doing rc5-32/12 cbc for 3s on 256 size blocks: 2369341 rc5-32/12 cbc's in 3.00s
Doing rc5-32/12 cbc for 3s on 1024 size blocks: 595638 rc5-32/12 cbc's in 3.00s
Doing rc5-32/12 cbc for 3s on 8192 size blocks: 74695 rc5-32/12 cbc's in 2.99s
Doing blowfish cbc for 3s on 16 size blocks: 19926444 blowfish cbc's in 3.00s
Doing blowfish cbc for 3s on 64 size blocks: 5309899 blowfish cbc's in 3.00s
Doing blowfish cbc for 3s on 256 size blocks: 1344396 blowfish cbc's in 3.00s
Doing blowfish cbc for 3s on 1024 size blocks: 338165 blowfish cbc's in 3.00s
Doing blowfish cbc for 3s on 8192 size blocks: 42403 blowfish cbc's in 3.00s
Doing cast cbc for 3s on 16 size blocks: 17925344 cast cbc's in 3.00s
Doing cast cbc for 3s on 64 size blocks: 4708134 cast cbc's in 3.00s
Doing cast cbc for 3s on 256 size blocks: 1192467 cast cbc's in 3.00s
Doing cast cbc for 3s on 1024 size blocks: 297940 cast cbc's in 3.00s
Doing cast cbc for 3s on 8192 size blocks: 37334 cast cbc's in 3.00s
Doing 512 bit private rsa's for 10s: 186668 512 bit private RSA's in 9.99s
Doing 512 bit public rsa's for 10s: 2268118 512 bit public RSA's in 10.00s
Doing 1024 bit private rsa's for 10s: 53031 1024 bit private RSA's in 10.00s
Doing 1024 bit public rsa's for 10s: 816698 1024 bit public RSA's in 9.99s
Doing 2048 bit private rsa's for 10s: 7417 2048 bit private RSA's in 9.99s
Doing 2048 bit public rsa's for 10s: 239856 2048 bit public RSA's in 10.00s
Doing 4096 bit private rsa's for 10s: 1030 4096 bit private RSA's in 10.00s
Doing 4096 bit public rsa's for 10s: 64115 4096 bit public RSA's in 10.00s
Doing 512 bit sign dsa's for 10s: 180875 512 bit DSA signs in 10.00s
Doing 512 bit verify dsa's for 10s: 196926 512 bit DSA verify in 9.99s
Doing 1024 bit sign dsa's for 10s: 76386 1024 bit DSA signs in 10.00s
Doing 1024 bit verify dsa's for 10s: 69939 1024 bit DSA verify in 10.00s
Doing 2048 bit sign dsa's for 10s: 24125 2048 bit DSA signs in 10.00s
Doing 2048 bit verify dsa's for 10s: 20460 2048 bit DSA verify in 9.99s
Doing 160 bit sign ecdsa's for 10s: 157258 160 bit ECDSA signs in 10.00s 
Doing 160 bit verify ecdsa's for 10s: 44429 160 bit ECDSA verify in 10.00s
Doing 192 bit sign ecdsa's for 10s: 130238 192 bit ECDSA signs in 9.99s 
Doing 192 bit verify ecdsa's for 10s: 37226 192 bit ECDSA verify in 10.00s
Doing 224 bit sign ecdsa's for 10s: 103526 224 bit ECDSA signs in 10.00s 
Doing 224 bit verify ecdsa's for 10s: 26223 224 bit ECDSA verify in 10.00s
Doing 256 bit sign ecdsa's for 10s: 88532 256 bit ECDSA signs in 9.99s 
Doing 256 bit verify ecdsa's for 10s: 22365 256 bit ECDSA verify in 10.00s
Doing 384 bit sign ecdsa's for 10s: 45982 384 bit ECDSA signs in 9.99s 
Doing 384 bit verify ecdsa's for 10s: 10622 384 bit ECDSA verify in 10.00s
Doing 521 bit sign ecdsa's for 10s: 24305 521 bit ECDSA signs in 9.99s 
Doing 521 bit verify ecdsa's for 10s: 4955 521 bit ECDSA verify in 9.99s
Doing 163 bit sign ecdsa's for 10s: 51447 163 bit ECDSA signs in 10.00s 
Doing 163 bit verify ecdsa's for 10s: 18398 163 bit ECDSA verify in 9.99s
Doing 233 bit sign ecdsa's for 10s: 25591 233 bit ECDSA signs in 9.99s 
Doing 233 bit verify ecdsa's for 10s: 13758 233 bit ECDSA verify in 10.00s
Doing 283 bit sign ecdsa's for 10s: 16485 283 bit ECDSA signs in 10.00s 
Doing 283 bit verify ecdsa's for 10s: 7303 283 bit ECDSA verify in 9.99s
Doing 409 bit sign ecdsa's for 10s: 7043 409 bit ECDSA signs in 10.00s 
Doing 409 bit verify ecdsa's for 10s: 4120 409 bit ECDSA verify in 10.00s
Doing 571 bit sign ecdsa's for 10s: 3187 571 bit ECDSA signs in 9.99s 
Doing 571 bit verify ecdsa's for 10s: 1798 571 bit ECDSA verify in 10.00s
Doing 163 bit sign ecdsa's for 10s: 51180 163 bit ECDSA signs in 10.00s 
Doing 163 bit verify ecdsa's for 10s: 17611 163 bit ECDSA verify in 10.00s
Doing 233 bit sign ecdsa's for 10s: 25516 233 bit ECDSA signs in 10.00s 
Doing 233 bit verify ecdsa's for 10s: 13049 233 bit ECDSA verify in 9.99s
Doing 283 bit sign ecdsa's for 10s: 16535 283 bit ECDSA signs in 9.99s 
Doing 283 bit verify ecdsa's for 10s: 6867 283 bit ECDSA verify in 9.99s
Doing 409 bit sign ecdsa's for 10s: 7026 409 bit ECDSA signs in 10.00s 
Doing 409 bit verify ecdsa's for 10s: 3821 409 bit ECDSA verify in 10.00s
Doing 571 bit sign ecdsa's for 10s: 3220 571 bit ECDSA signs in 10.00s 
Doing 571 bit verify ecdsa's for 10s: 1654 571 bit ECDSA verify in 10.00s
Doing 160 bit  ecdh's for 10s: 54157 160-bit ECDH ops in 10.00s
Doing 192 bit  ecdh's for 10s: 46502 192-bit ECDH ops in 10.00s
Doing 224 bit  ecdh's for 10s: 32144 224-bit ECDH ops in 10.00s
Doing 256 bit  ecdh's for 10s: 28259 256-bit ECDH ops in 10.00s
Doing 384 bit  ecdh's for 10s: 12896 384-bit ECDH ops in 10.00s
Doing 521 bit  ecdh's for 10s: 5890 521-bit ECDH ops in 10.00s
Doing 163 bit  ecdh's for 10s: 35377 163-bit ECDH ops in 10.00s
Doing 233 bit  ecdh's for 10s: 26642 233-bit ECDH ops in 10.00s
Doing 283 bit  ecdh's for 10s: 14257 283-bit ECDH ops in 10.00s
Doing 409 bit  ecdh's for 10s: 8417 409-bit ECDH ops in 10.00s
Doing 571 bit  ecdh's for 10s: 3714 571-bit ECDH ops in 10.00s
Doing 163 bit  ecdh's for 10s: 36461 163-bit ECDH ops in 9.99s
Doing 233 bit  ecdh's for 10s: 24729 233-bit ECDH ops in 9.99s
Doing 283 bit  ecdh's for 10s: 14245 283-bit ECDH ops in 9.99s
Doing 409 bit  ecdh's for 10s: 7926 409-bit ECDH ops in 9.99s
Doing 571 bit  ecdh's for 10s: 3391 571-bit ECDH ops in 9.99s
OpenSSL 1.0.1o 12 Jun 2015
built on: Sat Jun 20 20:56:10 2015
options:bn(64,64) rc4(16x,int) des(idx,cisc,16,int) aes(partial) idea(int) blowfish(idx) 
compiler: x86_64-pc-linux-gnu-gcc -I. -I.. -I../include  -fPIC -DOPENSSL_PIC -DZLIB -DOPENSSL_THREADS -D_REENTRANT -DDSO_DLFCN -DHAVE_DLFCN_H -DL_ENDIAN -Wall -DOPENSSL_IA32_SSE2 -DOPENSSL_BN_ASM_MONT -DOPENSSL_BN_ASM_MONT5 -DOPENSSL_BN_ASM_GF2m -DSHA1_ASM -DSHA256_ASM -DSHA512_ASM -DMD5_ASM -DAES_ASM -DVPAES_ASM -DBSAES_ASM -DWHIRLPOOL_ASM -DGHASH_ASM -march=nocona -O2 -pipe -fno-strict-aliasing -Wa,--noexecstack
The 'numbers' are in 1000s of bytes per second processed.
type             16 bytes     64 bytes    256 bytes   1024 bytes   8192 bytes
md2                  0.00         0.00         0.00         0.00         0.00 
mdc2             12703.04k    13369.15k    13621.08k    13696.00k    13716.14k
md4              86245.04k   254275.69k   571499.01k   827607.38k   949097.81k
md5              62760.47k   177103.02k   381346.93k   531916.12k   601743.36k
hmac(md5)        50744.58k   152272.49k   347920.64k   514579.80k   598272.68k
sha1             71380.87k   192120.94k   402205.01k   555094.70k   625538.39k
rmd160           38446.01k    89245.67k   158964.14k   197642.92k   212822.70k
rc4             373730.44k   628652.74k   732574.98k   765906.60k   772814.17k
des cbc          61057.80k    63338.96k    63540.74k    63550.46k    63750.14k
des ede3         23664.95k    24005.08k    24033.45k    24160.94k    24076.29k
idea cbc         76671.46k    79252.05k    79937.37k    80035.50k    80366.25k
seed cbc         63622.74k    64004.91k    64096.51k    64059.05k    64091.48k
rc2 cbc          37104.03k    37716.59k    38082.53k    38011.90k    38008.15k
rc5-32/12 cbc   178227.87k   196785.98k   202183.77k   203311.10k   204649.31k
blowfish cbc    106274.37k   113277.85k   114721.79k   115426.99k   115788.46k
cast cbc         95601.83k   100440.19k   101757.18k   101696.85k   101946.71k
aes-128 cbc     100296.84k   108067.97k   109795.93k   260763.48k   262496.26k
aes-192 cbc      84853.96k    88385.60k    90758.14k   218587.82k   221885.39k
aes-256 cbc      73251.68k    77044.18k    77933.48k   189125.29k   191463.42k
camellia-128 cbc    90980.81k   138102.49k   156191.25k   161385.13k   163348.48k
camellia-192 cbc    79106.17k   107921.02k   117938.86k   121656.32k   122866.35k
camellia-256 cbc    76959.42k   107975.00k   118350.85k   121749.16k   122738.01k
sha256           41240.62k    87622.04k   147071.49k   176615.08k   187449.34k
sha512           32263.84k   129692.68k   188145.15k   259924.99k   292009.30k
whirlpool        23503.99k    50129.49k    82917.12k    99129.69k   105706.94k
aes-128 ige     101861.45k   104384.39k   104530.69k   104812.20k   105026.90k
aes-192 ige      85719.06k    87179.58k    87444.74k    87874.90k    87919.27k
aes-256 ige      74281.56k    75227.43k    75467.52k    75453.78k    75587.58k
ghash           871658.65k  1409055.13k  1607221.33k  1650077.35k  1682341.89k
 sign    verify    sign/s verify/s
rsa  512 bits 0.000054s 0.000004s  18685.5 226811.8
rsa 1024 bits 0.000189s 0.000012s   5303.1  81751.6
rsa 2048 bits 0.001347s 0.000042s    742.4  23985.6
rsa 4096 bits 0.009709s 0.000156s    103.0   6411.5
 sign    verify    sign/s verify/s
dsa  512 bits 0.000055s 0.000051s  18087.5  19712.3
dsa 1024 bits 0.000131s 0.000143s   7638.6   6993.9
dsa 2048 bits 0.000415s 0.000488s   2412.5   2048.0
 sign    verify    sign/s verify/s
 160 bit ecdsa (secp160r1)   0.0001s   0.0002s  15725.8   4442.9
 192 bit ecdsa (nistp192)   0.0001s   0.0003s  13036.8   3722.6
 224 bit ecdsa (nistp224)   0.0001s   0.0004s  10352.6   2622.3
 256 bit ecdsa (nistp256)   0.0001s   0.0004s   8862.1   2236.5
 384 bit ecdsa (nistp384)   0.0002s   0.0009s   4602.8   1062.2
 521 bit ecdsa (nistp521)   0.0004s   0.0020s   2432.9    496.0
 163 bit ecdsa (nistk163)   0.0002s   0.0005s   5144.7   1841.6
 233 bit ecdsa (nistk233)   0.0004s   0.0007s   2561.7   1375.8
 283 bit ecdsa (nistk283)   0.0006s   0.0014s   1648.5    731.0
 409 bit ecdsa (nistk409)   0.0014s   0.0024s    704.3    412.0
 571 bit ecdsa (nistk571)   0.0031s   0.0056s    319.0    179.8
 163 bit ecdsa (nistb163)   0.0002s   0.0006s   5118.0   1761.1
 233 bit ecdsa (nistb233)   0.0004s   0.0008s   2551.6   1306.2
 283 bit ecdsa (nistb283)   0.0006s   0.0015s   1655.2    687.4
 409 bit ecdsa (nistb409)   0.0014s   0.0026s    702.6    382.1
 571 bit ecdsa (nistb571)   0.0031s   0.0060s    322.0    165.4
 op      op/s
 160 bit ecdh (secp160r1)   0.0002s   5415.7
 192 bit ecdh (nistp192)   0.0002s   4650.2
 224 bit ecdh (nistp224)   0.0003s   3214.4
 256 bit ecdh (nistp256)   0.0004s   2825.9
 384 bit ecdh (nistp384)   0.0008s   1289.6
 521 bit ecdh (nistp521)   0.0017s    589.0
 163 bit ecdh (nistk163)   0.0003s   3537.7
 233 bit ecdh (nistk233)   0.0004s   2664.2
 283 bit ecdh (nistk283)   0.0007s   1425.7
 409 bit ecdh (nistk409)   0.0012s    841.7
 571 bit ecdh (nistk571)   0.0027s    371.4
 163 bit ecdh (nistb163)   0.0003s   3649.7
 233 bit ecdh (nistb233)   0.0004s   2475.4
 283 bit ecdh (nistb283)   0.0007s   1425.9
 409 bit ecdh (nistb409)   0.0013s    793.4
 571 bit ecdh (nistb571)   0.0029s    339.4

# openssl speed -evp aes-256-cbc
Doing aes-256-cbc for 3s on 16 size blocks: 82075527 aes-256-cbc's in 3.00s
Doing aes-256-cbc for 3s on 64 size blocks: 21477090 aes-256-cbc's in 3.00s
Doing aes-256-cbc for 3s on 256 size blocks: 5430750 aes-256-cbc's in 2.99s
Doing aes-256-cbc for 3s on 1024 size blocks: 1363435 aes-256-cbc's in 3.00s
Doing aes-256-cbc for 3s on 8192 size blocks: 170434 aes-256-cbc's in 3.00s
OpenSSL 1.0.1o 12 Jun 2015
built on: Sat Jun 20 20:56:10 2015
options:bn(64,64) rc4(16x,int) des(idx,cisc,16,int) aes(partial) idea(int) blowfish(idx) 
compiler: x86_64-pc-linux-gnu-gcc -I. -I.. -I../include  -fPIC -DOPENSSL_PIC -DZLIB -DOPENSSL_THREADS -D_REENTRANT -DDSO_DLFCN -DHAVE_DLFCN_H -DL_ENDIAN -Wall -DOPENSSL_IA32_SSE2 -DOPENSSL_BN_ASM_MONT -DOPENSSL_BN_ASM_MONT5 -DOPENSSL_BN_ASM_GF2m -DSHA1_ASM -DSHA256_ASM -DSHA512_ASM -DMD5_ASM -DAES_ASM -DVPAES_ASM -DBSAES_ASM -DWHIRLPOOL_ASM -DGHASH_ASM -march=nocona -O2 -pipe -fno-strict-aliasing -Wa,--noexecstack
The 'numbers' are in 1000s of bytes per second processed.
type             16 bytes     64 bytes    256 bytes   1024 bytes   8192 bytes
aes-256-cbc     437736.14k   458177.92k   464973.91k   465385.81k   465398.44k

Atom C2750:

# openssl speed
# openssl speed
Doing mdc2 for 3s on 16 size blocks: 1237225 mdc2's in 3.00s
Doing mdc2 for 3s on 64 size blocks: 334859 mdc2's in 3.00s
Doing mdc2 for 3s on 256 size blocks: 85431 mdc2's in 3.00s
Doing mdc2 for 3s on 1024 size blocks: 21447 mdc2's in 3.00s
Doing mdc2 for 3s on 8192 size blocks: 2689 mdc2's in 3.00s
Doing md4 for 3s on 16 size blocks: 7341460 md4's in 3.00s
Doing md4 for 3s on 64 size blocks: 5809440 md4's in 3.00s
Doing md4 for 3s on 256 size blocks: 3495018 md4's in 3.00s
Doing md4 for 3s on 1024 size blocks: 1375859 md4's in 3.00s
Doing md4 for 3s on 8192 size blocks: 208156 md4's in 3.00s
Doing md5 for 3s on 16 size blocks: 5235418 md5's in 3.00s
Doing md5 for 3s on 64 size blocks: 4133535 md5's in 3.00s
Doing md5 for 3s on 256 size blocks: 2480694 md5's in 3.00s
Doing md5 for 3s on 1024 size blocks: 968874 md5's in 3.00s
Doing md5 for 3s on 8192 size blocks: 145509 md5's in 3.00s
Doing hmac(md5) for 3s on 16 size blocks: 4638823 hmac(md5)'s in 3.00s
Doing hmac(md5) for 3s on 64 size blocks: 3778913 hmac(md5)'s in 3.00s
Doing hmac(md5) for 3s on 256 size blocks: 2377536 hmac(md5)'s in 3.00s
Doing hmac(md5) for 3s on 1024 size blocks: 953963 hmac(md5)'s in 3.00s
Doing hmac(md5) for 3s on 8192 size blocks: 145376 hmac(md5)'s in 3.00s
Doing sha1 for 3s on 16 size blocks: 5084307 sha1's in 3.00s
Doing sha1 for 3s on 64 size blocks: 3571607 sha1's in 3.00s
Doing sha1 for 3s on 256 size blocks: 1900707 sha1's in 3.00s
Doing sha1 for 3s on 1024 size blocks: 661839 sha1's in 3.00s
Doing sha1 for 3s on 8192 size blocks: 93594 sha1's in 3.00s
Doing sha256 for 3s on 16 size blocks: 3619108 sha256's in 3.00s
Doing sha256 for 3s on 64 size blocks: 1952169 sha256's in 3.00s
Doing sha256 for 3s on 256 size blocks: 825413 sha256's in 3.00s
Doing sha256 for 3s on 1024 size blocks: 249514 sha256's in 3.00s
Doing sha256 for 3s on 8192 size blocks: 33217 sha256's in 3.00s
Doing sha512 for 3s on 16 size blocks: 2842737 sha512's in 3.00s
Doing sha512 for 3s on 64 size blocks: 2855728 sha512's in 3.00s
Doing sha512 for 3s on 256 size blocks: 1054445 sha512's in 3.00s
Doing sha512 for 3s on 1024 size blocks: 366700 sha512's in 3.00s
Doing sha512 for 3s on 8192 size blocks: 51829 sha512's in 3.00s
Doing whirlpool for 3s on 16 size blocks: 2592210 whirlpool's in 3.00s
Doing whirlpool for 3s on 64 size blocks: 1399727 whirlpool's in 3.00s
Doing whirlpool for 3s on 256 size blocks: 577815 whirlpool's in 3.00s
Doing whirlpool for 3s on 1024 size blocks: 173189 whirlpool's in 3.00s
Doing whirlpool for 3s on 8192 size blocks: 22967 whirlpool's in 3.00s
Doing rmd160 for 3s on 16 size blocks: 3719254 rmd160's in 3.00s
Doing rmd160 for 3s on 64 size blocks: 2316199 rmd160's in 3.00s
Doing rmd160 for 3s on 256 size blocks: 1090773 rmd160's in 3.00s
Doing rmd160 for 3s on 1024 size blocks: 350790 rmd160's in 3.00s
Doing rmd160 for 3s on 8192 size blocks: 47935 rmd160's in 3.00s
Doing rc4 for 3s on 16 size blocks: 32181778 rc4's in 3.00s
Doing rc4 for 3s on 64 size blocks: 13572012 rc4's in 3.00s
Doing rc4 for 3s on 256 size blocks: 4226247 rc4's in 3.00s
Doing rc4 for 3s on 1024 size blocks: 1131570 rc4's in 3.00s
Doing rc4 for 3s on 8192 size blocks: 144311 rc4's in 3.00s
Doing des cbc for 3s on 16 size blocks: 7464953 des cbc's in 3.00s
Doing des cbc for 3s on 64 size blocks: 1987444 des cbc's in 3.00s
Doing des cbc for 3s on 256 size blocks: 507095 des cbc's in 3.00s
Doing des cbc for 3s on 1024 size blocks: 127412 des cbc's in 3.00s
Doing des cbc for 3s on 8192 size blocks: 15940 des cbc's in 3.00s
Doing des ede3 for 3s on 16 size blocks: 2981222 des ede3's in 3.00s
Doing des ede3 for 3s on 64 size blocks: 762994 des ede3's in 3.00s
Doing des ede3 for 3s on 256 size blocks: 192406 des ede3's in 3.00s
Doing des ede3 for 3s on 1024 size blocks: 48204 des ede3's in 3.00s
Doing des ede3 for 3s on 8192 size blocks: 6028 des ede3's in 3.00s
Doing aes-128 cbc for 3s on 16 size blocks: 8191877 aes-128 cbc's in 3.00s
Doing aes-128 cbc for 3s on 64 size blocks: 2308084 aes-128 cbc's in 3.00s
Doing aes-128 cbc for 3s on 256 size blocks: 601530 aes-128 cbc's in 3.00s
Doing aes-128 cbc for 3s on 1024 size blocks: 376976 aes-128 cbc's in 3.00s
Doing aes-128 cbc for 3s on 8192 size blocks: 47988 aes-128 cbc's in 3.00s
Doing aes-192 cbc for 3s on 16 size blocks: 6996169 aes-192 cbc's in 3.00s
Doing aes-192 cbc for 3s on 64 size blocks: 1935438 aes-192 cbc's in 3.00s
Doing aes-192 cbc for 3s on 256 size blocks: 500913 aes-192 cbc's in 3.00s
Doing aes-192 cbc for 3s on 1024 size blocks: 319605 aes-192 cbc's in 3.00s
Doing aes-192 cbc for 3s on 8192 size blocks: 40572 aes-192 cbc's in 3.00s
Doing aes-256 cbc for 3s on 16 size blocks: 6033048 aes-256 cbc's in 3.00s
Doing aes-256 cbc for 3s on 64 size blocks: 1656115 aes-256 cbc's in 3.00s
Doing aes-256 cbc for 3s on 256 size blocks: 428478 aes-256 cbc's in 3.00s
Doing aes-256 cbc for 3s on 1024 size blocks: 279083 aes-256 cbc's in 3.00s
Doing aes-256 cbc for 3s on 8192 size blocks: 35165 aes-256 cbc's in 3.00s
Doing aes-128 ige for 3s on 16 size blocks: 8407668 aes-128 ige's in 3.00s
Doing aes-128 ige for 3s on 64 size blocks: 2259882 aes-128 ige's in 3.00s
Doing aes-128 ige for 3s on 256 size blocks: 577873 aes-128 ige's in 3.00s
Doing aes-128 ige for 3s on 1024 size blocks: 145298 aes-128 ige's in 3.00s
Doing aes-128 ige for 3s on 8192 size blocks: 18185 aes-128 ige's in 3.00s
Doing aes-192 ige for 3s on 16 size blocks: 7159765 aes-192 ige's in 3.00s
Doing aes-192 ige for 3s on 64 size blocks: 1901773 aes-192 ige's in 3.00s
Doing aes-192 ige for 3s on 256 size blocks: 484446 aes-192 ige's in 3.00s
Doing aes-192 ige for 3s on 1024 size blocks: 121686 aes-192 ige's in 3.00s
Doing aes-192 ige for 3s on 8192 size blocks: 15225 aes-192 ige's in 3.00s
Doing aes-256 ige for 3s on 16 size blocks: 6150421 aes-256 ige's in 3.00s
Doing aes-256 ige for 3s on 64 size blocks: 1625242 aes-256 ige's in 3.00s
Doing aes-256 ige for 3s on 256 size blocks: 415973 aes-256 ige's in 3.00s
Doing aes-256 ige for 3s on 1024 size blocks: 104607 aes-256 ige's in 3.00s
Doing aes-256 ige for 3s on 8192 size blocks: 13093 aes-256 ige's in 3.00s
Doing ghash for 3s on 16 size blocks: 55986826 ghash's in 3.00s
Doing ghash for 3s on 64 size blocks: 26263878 ghash's in 3.00s
Doing ghash for 3s on 256 size blocks: 8335308 ghash's in 3.00s
Doing ghash for 3s on 1024 size blocks: 2238027 ghash's in 3.00s
Doing ghash for 3s on 8192 size blocks: 286956 ghash's in 3.00s
Doing camellia-128 cbc for 3s on 16 size blocks: 7978113 camellia-128 cbc's in 3.00s
Doing camellia-128 cbc for 3s on 64 size blocks: 3555926 camellia-128 cbc's in 3.00s
Doing camellia-128 cbc for 3s on 256 size blocks: 1104288 camellia-128 cbc's in 3.00s
Doing camellia-128 cbc for 3s on 1024 size blocks: 294306 camellia-128 cbc's in 3.00s
Doing camellia-128 cbc for 3s on 8192 size blocks: 37493 camellia-128 cbc's in 3.00s
Doing camellia-192 cbc for 3s on 16 size blocks: 7046782 camellia-192 cbc's in 3.00s
Doing camellia-192 cbc for 3s on 64 size blocks: 2862624 camellia-192 cbc's in 3.00s
Doing camellia-192 cbc for 3s on 256 size blocks: 854300 camellia-192 cbc's in 3.00s
Doing camellia-192 cbc for 3s on 1024 size blocks: 224337 camellia-192 cbc's in 3.00s
Doing camellia-192 cbc for 3s on 8192 size blocks: 28444 camellia-192 cbc's in 3.00s
Doing camellia-256 cbc for 3s on 16 size blocks: 7040128 camellia-256 cbc's in 3.00s
Doing camellia-256 cbc for 3s on 64 size blocks: 2860326 camellia-256 cbc's in 3.00s
Doing camellia-256 cbc for 3s on 256 size blocks: 854188 camellia-256 cbc's in 3.00s
Doing camellia-256 cbc for 3s on 1024 size blocks: 224316 camellia-256 cbc's in 3.00s
Doing camellia-256 cbc for 3s on 8192 size blocks: 28439 camellia-256 cbc's in 3.00s
Doing idea cbc for 3s on 16 size blocks: 7727576 idea cbc's in 3.00s
Doing idea cbc for 3s on 64 size blocks: 2110967 idea cbc's in 3.00s
Doing idea cbc for 3s on 256 size blocks: 540105 idea cbc's in 3.00s
Doing idea cbc for 3s on 1024 size blocks: 135818 idea cbc's in 3.00s
Doing idea cbc for 3s on 8192 size blocks: 16999 idea cbc's in 3.00s
Doing seed cbc for 3s on 16 size blocks: 8531935 seed cbc's in 3.00s
Doing seed cbc for 3s on 64 size blocks: 2281697 seed cbc's in 3.00s
Doing seed cbc for 3s on 256 size blocks: 579778 seed cbc's in 3.00s
Doing seed cbc for 3s on 1024 size blocks: 145722 seed cbc's in 3.00s
Doing seed cbc for 3s on 8192 size blocks: 18231 seed cbc's in 3.00s
Doing rc2 cbc for 3s on 16 size blocks: 5232519 rc2 cbc's in 2.99s
Doing rc2 cbc for 3s on 64 size blocks: 1390420 rc2 cbc's in 3.00s
Doing rc2 cbc for 3s on 256 size blocks: 352366 rc2 cbc's in 3.00s
Doing rc2 cbc for 3s on 1024 size blocks: 88388 rc2 cbc's in 3.00s
Doing rc2 cbc for 3s on 8192 size blocks: 11058 rc2 cbc's in 3.00s
Doing rc5-32/12 cbc for 3s on 16 size blocks: 17960647 rc5-32/12 cbc's in 3.00s
Doing rc5-32/12 cbc for 3s on 64 size blocks: 5324175 rc5-32/12 cbc's in 3.00s
Doing rc5-32/12 cbc for 3s on 256 size blocks: 1411637 rc5-32/12 cbc's in 3.00s
Doing rc5-32/12 cbc for 3s on 1024 size blocks: 358518 rc5-32/12 cbc's in 3.00s
Doing rc5-32/12 cbc for 3s on 8192 size blocks: 45012 rc5-32/12 cbc's in 3.00s
Doing blowfish cbc for 3s on 16 size blocks: 13639956 blowfish cbc's in 3.00s
Doing blowfish cbc for 3s on 64 size blocks: 4008723 blowfish cbc's in 3.00s
Doing blowfish cbc for 3s on 256 size blocks: 1045847 blowfish cbc's in 3.00s
Doing blowfish cbc for 3s on 1024 size blocks: 264920 blowfish cbc's in 3.00s
Doing blowfish cbc for 3s on 8192 size blocks: 33216 blowfish cbc's in 3.00s
Doing cast cbc for 3s on 16 size blocks: 11918048 cast cbc's in 3.00s
Doing cast cbc for 3s on 64 size blocks: 3333624 cast cbc's in 3.00s
Doing cast cbc for 3s on 256 size blocks: 866433 cast cbc's in 3.00s
Doing cast cbc for 3s on 1024 size blocks: 218654 cast cbc's in 3.00s
Doing cast cbc for 3s on 8192 size blocks: 27397 cast cbc's in 3.00s
Doing 512 bit private rsa's for 10s: 55770 512 bit private RSA's in 10.00s
Doing 512 bit public rsa's for 10s: 699257 512 bit public RSA's in 10.00s
Doing 1024 bit private rsa's for 10s: 15853 1024 bit private RSA's in 10.00s
Doing 1024 bit public rsa's for 10s: 236209 1024 bit public RSA's in 10.00s
Doing 2048 bit private rsa's for 10s: 2094 2048 bit private RSA's in 10.00s
Doing 2048 bit public rsa's for 10s: 68809 2048 bit public RSA's in 10.00s
Doing 4096 bit private rsa's for 10s: 289 4096 bit private RSA's in 10.02s
Doing 4096 bit public rsa's for 10s: 18204 4096 bit public RSA's in 10.00s
Doing 512 bit sign dsa's for 10s: 59496 512 bit DSA signs in 10.00s
Doing 512 bit verify dsa's for 10s: 59516 512 bit DSA verify in 10.00s
Doing 1024 bit sign dsa's for 10s: 22856 1024 bit DSA signs in 10.00s
Doing 1024 bit verify dsa's for 10s: 19780 1024 bit DSA verify in 10.00s
Doing 2048 bit sign dsa's for 10s: 6863 2048 bit DSA signs in 10.00s
Doing 2048 bit verify dsa's for 10s: 5667 2048 bit DSA verify in 10.00s
Doing 160 bit sign ecdsa's for 10s: 59376 160 bit ECDSA signs in 10.00s 
Doing 160 bit verify ecdsa's for 10s: 15834 160 bit ECDSA verify in 10.00s
Doing 192 bit sign ecdsa's for 10s: 49930 192 bit ECDSA signs in 10.00s 
Doing 192 bit verify ecdsa's for 10s: 13078 192 bit ECDSA verify in 10.00s
Doing 224 bit sign ecdsa's for 10s: 39138 224 bit ECDSA signs in 10.00s 
Doing 224 bit verify ecdsa's for 10s: 9266 224 bit ECDSA verify in 10.00s
Doing 256 bit sign ecdsa's for 10s: 33863 256 bit ECDSA signs in 10.00s 
Doing 256 bit verify ecdsa's for 10s: 7956 256 bit ECDSA verify in 10.00s
Doing 384 bit sign ecdsa's for 10s: 16944 384 bit ECDSA signs in 10.00s 
Doing 384 bit verify ecdsa's for 10s: 3548 384 bit ECDSA verify in 10.00s
Doing 521 bit sign ecdsa's for 10s: 8331 521 bit ECDSA signs in 10.00s 
Doing 521 bit verify ecdsa's for 10s: 1557 521 bit ECDSA verify in 10.00s
Doing 163 bit sign ecdsa's for 10s: 23619 163 bit ECDSA signs in 10.00s 
Doing 163 bit verify ecdsa's for 10s: 7744 163 bit ECDSA verify in 10.00s
Doing 233 bit sign ecdsa's for 10s: 12041 233 bit ECDSA signs in 10.00s 
Doing 233 bit verify ecdsa's for 10s: 5702 233 bit ECDSA verify in 10.00s
Doing 283 bit sign ecdsa's for 10s: 7712 283 bit ECDSA signs in 10.00s 
Doing 283 bit verify ecdsa's for 10s: 3185 283 bit ECDSA verify in 10.00s
Doing 409 bit sign ecdsa's for 10s: 3373 409 bit ECDSA signs in 10.01s 
Doing 409 bit verify ecdsa's for 10s: 1770 409 bit ECDSA verify in 10.00s
Doing 571 bit sign ecdsa's for 10s: 1534 571 bit ECDSA signs in 10.01s 
Doing 571 bit verify ecdsa's for 10s: 796 571 bit ECDSA verify in 10.01s
Doing 163 bit sign ecdsa's for 10s: 23857 163 bit ECDSA signs in 10.00s 
Doing 163 bit verify ecdsa's for 10s: 7313 163 bit ECDSA verify in 10.00s
Doing 233 bit sign ecdsa's for 10s: 12065 233 bit ECDSA signs in 10.00s 
Doing 233 bit verify ecdsa's for 10s: 5468 233 bit ECDSA verify in 10.00s
Doing 283 bit sign ecdsa's for 10s: 7698 283 bit ECDSA signs in 10.00s 
Doing 283 bit verify ecdsa's for 10s: 3015 283 bit ECDSA verify in 10.00s
Doing 409 bit sign ecdsa's for 10s: 3369 409 bit ECDSA signs in 10.00s 
Doing 409 bit verify ecdsa's for 10s: 1643 409 bit ECDSA verify in 10.00s
Doing 571 bit sign ecdsa's for 10s: 1532 571 bit ECDSA signs in 10.00s 
Doing 571 bit verify ecdsa's for 10s: 732 571 bit ECDSA verify in 10.00s
Doing 160 bit  ecdh's for 10s: 19662 160-bit ECDH ops in 10.00s
Doing 192 bit  ecdh's for 10s: 15458 192-bit ECDH ops in 10.00s
Doing 224 bit  ecdh's for 10s: 11125 224-bit ECDH ops in 10.00s
Doing 256 bit  ecdh's for 10s: 9843 256-bit ECDH ops in 10.00s
Doing 384 bit  ecdh's for 10s: 4306 384-bit ECDH ops in 10.00s
Doing 521 bit  ecdh's for 10s: 1868 521-bit ECDH ops in 10.01s
Doing 163 bit  ecdh's for 10s: 15779 163-bit ECDH ops in 10.00s
Doing 233 bit  ecdh's for 10s: 11886 233-bit ECDH ops in 10.00s
Doing 283 bit  ecdh's for 10s: 6514 283-bit ECDH ops in 10.00s
Doing 409 bit  ecdh's for 10s: 3609 409-bit ECDH ops in 10.00s
Doing 571 bit  ecdh's for 10s: 1630 571-bit ECDH ops in 10.00s
Doing 163 bit  ecdh's for 10s: 15227 163-bit ECDH ops in 10.00s
Doing 233 bit  ecdh's for 10s: 11241 233-bit ECDH ops in 10.00s
Doing 283 bit  ecdh's for 10s: 6180 283-bit ECDH ops in 10.00s
Doing 409 bit  ecdh's for 10s: 3361 409-bit ECDH ops in 10.00s
Doing 571 bit  ecdh's for 10s: 1506 571-bit ECDH ops in 10.01s
OpenSSL 1.0.1o 12 Jun 2015
built on: Mon Jul  6 22:23:47 2015
options:bn(64,64) rc4(16x,int) des(idx,cisc,16,int) aes(partial) idea(int) blowfish(idx) 
compiler: x86_64-pc-linux-gnu-gcc -I. -I.. -I../include  -fPIC -DOPENSSL_PIC -DZLIB -DOPENSSL_THREADS -D_REENTRANT -DDSO_DLFCN -DHAVE_DLFCN_H -DL_ENDIAN -Wall -DOPENSSL_IA32_SSE2 -DOPENSSL_BN_ASM_MONT -DOPENSSL_BN_ASM_MONT5 -DOPENSSL_BN_ASM_GF2m -DSHA1_ASM -DSHA256_ASM -DSHA512_ASM -DMD5_ASM -DAES_ASM -DVPAES_ASM -DBSAES_ASM -DWHIRLPOOL_ASM -DGHASH_ASM -march=nocona -O2 -pipe -fno-strict-aliasing -Wa,--noexecstack
The 'numbers' are in 1000s of bytes per second processed.
type             16 bytes     64 bytes    256 bytes   1024 bytes   8192 bytes
md2                  0.00         0.00         0.00         0.00         0.00 
mdc2              6598.53k     7143.66k     7290.11k     7320.58k     7342.76k
md4              39154.45k   123934.72k   298241.54k   469626.54k   568404.65k
md5              27922.23k    88182.08k   211685.89k   330708.99k   397336.58k
hmac(md5)        24740.39k    80616.81k   202883.07k   325619.37k   396973.40k
sha1             27116.30k    76194.28k   162193.66k   225907.71k   255574.02k
rmd160           19836.02k    49412.25k    93079.30k   119736.32k   130894.51k
rc4             171636.15k   289536.26k   360639.74k   386242.56k   394065.24k
des cbc          39813.08k    42398.81k    43272.11k    43489.96k    43526.83k
des ede3         15899.85k    16277.21k    16418.65k    16453.63k    16460.46k
idea cbc         41213.74k    45033.96k    46088.96k    46359.21k    46418.60k
seed cbc         45503.65k    48676.20k    49474.39k    49739.78k    49782.78k
rc2 cbc          28000.10k    29662.29k    30068.57k    30169.77k    30195.71k
rc5-32/12 cbc    95790.12k   113582.40k   120459.69k   122374.14k   122912.77k
blowfish cbc     72746.43k    85519.42k    89245.61k    90426.03k    90701.82k
cast cbc         63562.92k    71117.31k    73935.62k    74633.90k    74812.07k
aes-128 cbc      43690.01k    49239.13k    51330.56k   128674.47k   131039.23k
aes-192 cbc      37312.90k    41289.34k    42744.58k   109091.84k   110788.61k
aes-256 cbc      32176.26k    35330.45k    36563.46k    95260.33k    96023.89k
camellia-128 cbc    42549.94k    75859.75k    94232.58k   100456.45k   102380.89k
camellia-192 cbc    37582.84k    61069.31k    72900.27k    76573.70k    77671.08k
camellia-256 cbc    37547.35k    61020.29k    72890.71k    76566.53k    77657.43k
sha256           19301.91k    41646.27k    70435.24k    85167.45k    90704.55k
sha512           15161.26k    60922.20k    89979.31k   125166.93k   141527.72k
whirlpool        13825.12k    29860.84k    49306.88k    59115.18k    62715.22k
aes-128 ige      44840.90k    48210.82k    49311.83k    49595.05k    49657.17k
aes-192 ige      38185.41k    40571.16k    41339.39k    41535.49k    41574.40k
aes-256 ige      32802.25k    34671.83k    35496.36k    35705.86k    35752.62k
ghash           298596.41k   560296.06k   711279.62k   763913.22k   783581.18k
 sign    verify    sign/s verify/s
rsa  512 bits 0.000179s 0.000014s   5577.0  69925.7
rsa 1024 bits 0.000631s 0.000042s   1585.3  23620.9
rsa 2048 bits 0.004776s 0.000145s    209.4   6880.9
rsa 4096 bits 0.034671s 0.000549s     28.8   1820.4
 sign    verify    sign/s verify/s
dsa  512 bits 0.000168s 0.000168s   5949.6   5951.6
dsa 1024 bits 0.000438s 0.000506s   2285.6   1978.0
dsa 2048 bits 0.001457s 0.001765s    686.3    566.7
 sign    verify    sign/s verify/s
 160 bit ecdsa (secp160r1)   0.0002s   0.0006s   5937.6   1583.4
 192 bit ecdsa (nistp192)   0.0002s   0.0008s   4993.0   1307.8
 224 bit ecdsa (nistp224)   0.0003s   0.0011s   3913.8    926.6
 256 bit ecdsa (nistp256)   0.0003s   0.0013s   3386.3    795.6
 384 bit ecdsa (nistp384)   0.0006s   0.0028s   1694.4    354.8
 521 bit ecdsa (nistp521)   0.0012s   0.0064s    833.1    155.7
 163 bit ecdsa (nistk163)   0.0004s   0.0013s   2361.9    774.4
 233 bit ecdsa (nistk233)   0.0008s   0.0018s   1204.1    570.2
 283 bit ecdsa (nistk283)   0.0013s   0.0031s    771.2    318.5
 409 bit ecdsa (nistk409)   0.0030s   0.0056s    337.0    177.0
 571 bit ecdsa (nistk571)   0.0065s   0.0126s    153.2     79.5
 163 bit ecdsa (nistb163)   0.0004s   0.0014s   2385.7    731.3
 233 bit ecdsa (nistb233)   0.0008s   0.0018s   1206.5    546.8
 283 bit ecdsa (nistb283)   0.0013s   0.0033s    769.8    301.5
 409 bit ecdsa (nistb409)   0.0030s   0.0061s    336.9    164.3
 571 bit ecdsa (nistb571)   0.0065s   0.0137s    153.2     73.2
 op      op/s
 160 bit ecdh (secp160r1)   0.0005s   1966.2
 192 bit ecdh (nistp192)   0.0006s   1545.8
 224 bit ecdh (nistp224)   0.0009s   1112.5
 256 bit ecdh (nistp256)   0.0010s    984.3
 384 bit ecdh (nistp384)   0.0023s    430.6
 521 bit ecdh (nistp521)   0.0054s    186.6
 163 bit ecdh (nistk163)   0.0006s   1577.9
 233 bit ecdh (nistk233)   0.0008s   1188.6
 283 bit ecdh (nistk283)   0.0015s    651.4
 409 bit ecdh (nistk409)   0.0028s    360.9
 571 bit ecdh (nistk571)   0.0061s    163.0
 163 bit ecdh (nistb163)   0.0007s   1522.7
 233 bit ecdh (nistb233)   0.0009s   1124.1
 283 bit ecdh (nistb283)   0.0016s    618.0
 409 bit ecdh (nistb409)   0.0030s    336.1
 571 bit ecdh (nistb571)   0.0066s    150.4

# openssl speed -evp aes-256-cbc
Doing aes-256-cbc for 3s on 16 size blocks: 35839602 aes-256-cbc's in 3.00s
Doing aes-256-cbc for 3s on 64 size blocks: 12281656 aes-256-cbc's in 3.00s
Doing aes-256-cbc for 3s on 256 size blocks: 3548960 aes-256-cbc's in 3.00s
Doing aes-256-cbc for 3s on 1024 size blocks: 923282 aes-256-cbc's in 3.00s
Doing aes-256-cbc for 3s on 8192 size blocks: 116726 aes-256-cbc's in 3.00s
OpenSSL 1.0.1o 12 Jun 2015
built on: Mon Jul  6 22:23:47 2015
options:bn(64,64) rc4(16x,int) des(idx,cisc,16,int) aes(partial) idea(int) blowfish(idx) 
compiler: x86_64-pc-linux-gnu-gcc -I. -I.. -I../include  -fPIC -DOPENSSL_PIC -DZLIB -DOPENSSL_THREADS -D_REENTRANT -DDSO_DLFCN -DHAVE_DLFCN_H -DL_ENDIAN -Wall -DOPENSSL_IA32_SSE2 -DOPENSSL_BN_ASM_MONT -DOPENSSL_BN_ASM_MONT5 -DOPENSSL_BN_ASM_GF2m -DSHA1_ASM -DSHA256_ASM -DSHA512_ASM -DMD5_ASM -DAES_ASM -DVPAES_ASM -DBSAES_ASM -DWHIRLPOOL_ASM -DGHASH_ASM -march=nocona -O2 -pipe -fno-strict-aliasing -Wa,--noexecstack
The 'numbers' are in 1000s of bytes per second processed.
type             16 bytes     64 bytes    256 bytes   1024 bytes   8192 bytes
aes-256-cbc     191144.54k   262008.66k   302844.59k   315146.92k   318739.80k

Zaujalo mě, že šifrování větších bloků pomocí HW akcelerace se mezi oběma procesory liší znatelně méně než bez ní.

No a nakonec jsem vyzkoušel jednovláknové kódování devatenácti minut zvuku z WAVu do MP3 pomocí Lame.

i5-2400:

# time lame --preset standard Hergest\ Ridge\ \(Part\ One\).wav
...
real    0m23.049s
user    0m22.896s
sys 0m0.129s

Atom C2750:

# time lame --preset standard Hergest\ Ridge\ \(Part\ One\).wav
...
real    1m1.324s
user    1m0.325s
sys     0m0.356s

Závěr nechť si udělá každý sám...:-)

       

Hodnocení: 62 %

        špatné • dobré        

Tiskni Sdílej: Linkuj Jaggni to Vybrali.sme.sk Google Del.icio.us Facebook

Komentáře

Nástroje: Začni sledovat (1) ?Zašle upozornění na váš email při vložení nového komentáře. , Tisk

Vložit další komentář

10.7.2015 06:43 Miriam | skóre: 3 | blog: zivot
Rozbalit Rozbalit vše Re: Srovnání i5-2400@3.10GHz s Atomem C2750@2.40GHz
Odpovědět | Sbalit | Link | Blokovat | Admin
Ty vole, proč si má udělat každý závěr sám? Kvůli tomu jsem to otvírala.
10.7.2015 09:32 Honza Jaroš | skóre: 6 | blog: moje_strana_plotu | Bohnice
Rozbalit Rozbalit vše Re: Srovnání i5-2400@3.10GHz s Atomem C2750@2.40GHz
Ty krávo, protože někomu může vadit poloviční výkon v jednom vláknu a někdo jiný může být spokojený se stejným výkonem s podstatně menší spotřebou při využití všech jader.
10.7.2015 09:47 marek_hb
Rozbalit Rozbalit vše Re: Srovnání i5-2400@3.10GHz s Atomem C2750@2.40GHz
to je ten kráva! :-) - o to je to celé veselejší :)
10.7.2015 10:00 Miriam | skóre: 3 | blog: zivot
Rozbalit Rozbalit vše Re: Srovnání i5-2400@3.10GHz s Atomem C2750@2.40GHz
Dík, to se mi líbí! Skvělý zápisek na tuhle žumpu, kdybys to trochu učesal, nejspíš by se to dalo publikovat na diitu nebo rootu jako článek.
10.7.2015 19:58 Víla Simona
Rozbalit Rozbalit vše Re: Srovnání i5-2400@3.10GHz s Atomem C2750@2.40GHz
Tak si se stal(a) oficiální členkou kravího stáda, vole. :D.
Agent avatar 10.7.2015 22:32 Agent | blog: Life_in_Pieces | HC city
Rozbalit Rozbalit vše Re: Srovnání i5-2400@3.10GHz s Atomem C2750@2.40GHz
Žumpu z toho děláš ty. Až odejdeš, žumpa tu nebude.
Nevěděl zpočátku, co si počít, jak žít, co dělat, ale brzy se vpravil do role samotáře.
10.7.2015 07:28 Michal Kubeček | skóre: 72 | Luštěnice
Rozbalit Rozbalit vše Re: Srovnání i5-2400@3.10GHz s Atomem C2750@2.40GHz
Odpovědět | Sbalit | Link | Blokovat | Admin
Před samotnou kompilací jsem pro jistotu spustil 'tar -cp linux-4.0.5-gentoo | cat > /dev/null', aby toho bylo co nejvíc v cachi a tudíž aby byl proces co nejméně zpomalován přístupem na disk.

Nebylo by jednodušší to rovnou překládat na tmpfs, aby to disk neovlivňoval vůbec? (A k čemu je tam dobrý ten cat?)

Vzhledem k tomu, že jste to spouštěl ve virtuálu, což znamená dva operační systémy, kde mohly další procesy náhodně ovlivňovat výsledky, bylo by vhodné aspoň spustit každý test víckrát a uvést aspoň průměr a směrodatnou odchylku.

Ohledně prezentace: je to strašně nepřehledné. Doporučoval bych do textu dát jen tabulku s výsledky jednotlivých testů a kompletní výstupy přiložit zvlášť, aby se do nich čtenář podíval až když ho budou zajímat detaily.

10.7.2015 09:28 Honza Jaroš | skóre: 6 | blog: moje_strana_plotu | Bohnice
Rozbalit Rozbalit vše Re: Srovnání i5-2400@3.10GHz s Atomem C2750@2.40GHz
Na rovinu - dělal jsem to primárně pro sebe, uvažoval jsem o tom, jestli to vůbec má cenu sem dávat. Asi nemělo...
10.7.2015 10:00 Miriam | skóre: 3 | blog: zivot
Rozbalit Rozbalit vše Re: Srovnání i5-2400@3.10GHz s Atomem C2750@2.40GHz
Mělo, já jsem ti vděčná za to, že jsi to sdílel.
10.7.2015 10:16 Michal Kubeček | skóre: 72 | Luštěnice
Rozbalit Rozbalit vše Re: Srovnání i5-2400@3.10GHz s Atomem C2750@2.40GHz
Určitě mělo, to porovnání je rozhodně zajímavé. Jen jsem upozornil na některé věci, které se daly udělat lépe. Hlavně ta prezentace výsledků, zorientovat se ve výstupech jednotlivých příkazů a vytáhnout z nich to podstatné chvíli trvá; kdyby tam byla přehledná tabulka a pak zvlášť detailní logy, bylo by to mnohem lepší.
10.7.2015 10:38 Honza Jaroš | skóre: 6 | blog: moje_strana_plotu | Bohnice
Rozbalit Rozbalit vše Re: Srovnání i5-2400@3.10GHz s Atomem C2750@2.40GHz
Jasně. Problém je v tom, že to znamená další plus mínus hodinu, co bych tomu musel věnovat. A nejsem si jistý, jestli mi to za to stojí.
10.7.2015 09:44 Honza Jaroš | skóre: 6 | blog: moje_strana_plotu | Bohnice
Rozbalit Rozbalit vše Re: Srovnání i5-2400@3.10GHz s Atomem C2750@2.40GHz
Jinak na těch "dvou systémech" v danou chvíli běžel v podstatě jen rsyslog a SSH (+QEMU bez jakýchkoli dalších nástrojů na fyzickém stroji). Vícekrát jsem spouštěl sysbench a OpenSSL a lišily se v podstatě jen o desetiny vteřiny.
10.7.2015 10:21 Michal Kubeček | skóre: 72 | Luštěnice
Rozbalit Rozbalit vše Re: Srovnání i5-2400@3.10GHz s Atomem C2750@2.40GHz

Tohle je na dnešních systémech bohužel čím dál méně pravda. Ještě tak před 7-8 lety jsem na školeních ukazoval výstup ps a komentoval, že přestože systém (s KDE) běží už několik hodin, jsou tam jen jeden nebo dva procesy, které nasbíraly přes sekundu času CPU (a žádný přes dvě). Dnes už by to vypadalo výrazně jinak.

Když jsou výsledky tak zřetelně odlišné, jako je to ve většině testů tady, tak to moc nevadí. Pokud ale někdy budete dělat test, kde budou jen malé rozdíly mezi dvěma variantami, může to výsledky dost ovlivnit.

10.7.2015 10:30 Honza Jaroš | skóre: 6 | blog: moje_strana_plotu | Bohnice
Rozbalit Rozbalit vše Re: Srovnání i5-2400@3.10GHz s Atomem C2750@2.40GHz
Teď si nejsem jistý, co není pravda. Je tam základní Gentoo a ve výpisech procesů opravdu vidím jen init (OpenRC), rsyslogd, sshd a udev, na fyzickém stroji ještě qemu-system-x86. Samozřejmě na takovém Ubuntu by to vypadalo jinak.
10.7.2015 10:40 Honza Jaroš | skóre: 6 | blog: moje_strana_plotu | Bohnice
Rozbalit Rozbalit vše Re: Srovnání i5-2400@3.10GHz s Atomem C2750@2.40GHz
A co se týče posledního testu, tak s nastartovaným LXC vypadá výpis z pstree takhle: 
init─┬─6*[agetty]
     ├─lxc-monitord
     ├─lxc-start───init─┬─dbus-daemon
     │                  └─5*[getty]
     ├─rsyslogd─┬─{in:imklog}
     │          ├─{in:imuxsock}
     │          └─{rs:main Q:Reg}
     ├─sshd───sshd───bash───pstree
     └─systemd-udevd
10.7.2015 11:14 Michal Kubeček | skóre: 72 | Luštěnice
Rozbalit Rozbalit vše Re: Srovnání i5-2400@3.10GHz s Atomem C2750@2.40GHz
Tak to pak ano, ale takhle úsporné systémy jsou bohužel čím dál vzácnější.
10.7.2015 11:21 Honza Jaroš | skóre: 6 | blog: moje_strana_plotu | Bohnice
Rozbalit Rozbalit vše Re: Srovnání i5-2400@3.10GHz s Atomem C2750@2.40GHz
No já už právě kvůli tomu čím dál častěji používám Gentoo, protože to je stále ještě relativně čisté. A pokud nepotřebuju poslední verze balíčků, tak CentOS6, ten tu s námi ještě pár let bude a při instalaci bez instalátoru člověk také dostane relativně čistý systém. Co budu dělat, až i tyhle poslední ostrůvky zmizí, zatím nevím...
10.7.2015 17:13 Tomeši, Tomeši, ty bejku jeden
Rozbalit Rozbalit vše Re: Srovnání i5-2400@3.10GHz s Atomem C2750@2.40GHz
Co budu dělat, až i tyhle poslední ostrůvky zmizí, zatím nevím...
Oběsíš se.
10.7.2015 11:53 Kvakor
Rozbalit Rozbalit vše Re: Srovnání i5-2400@3.10GHz s Atomem C2750@2.40GHz
Před samotnou kompilací jsem pro jistotu spustil 'tar -cp linux-4.0.5-gentoo | cat > /dev/null', aby toho bylo co nejvíc v cachi a tudíž aby byl proces co nejméně zpomalován přístupem na disk.

Nebylo by jednodušší to rovnou překládat na tmpfs, aby to disk neovlivňoval vůbec?

Pokud je málo paměti, tak i data na tmpfs mohou být odswapována, takže pokud se dělají podobné testy, tak je vhodné před testem swap vypnot. Případně se může místo tmpfs použít ramfs, ten je v jádrech přítoment automaticky (viz cat /proc/filesystems), ale nemá kvóty, limity a hlavně možnost odlložit data do swapu. Tudíž je velmi snadné jeho pomocí dostat systém do stavu, kdy dojde paměť a jádro s tím nemůže nic udělat, navíc se paměť zabraná ramfs (i tmpfs) jeví jako disková cache, tudíž ji mnoho "paměťometrů" ignoruje.
10.7.2015 15:20 Ondrej 'SanTiago' Zajicek
Rozbalit Rozbalit vše Re: Srovnání i5-2400@3.10GHz s Atomem C2750@2.40GHz
Pokud je málo paměti, tak i data na tmpfs mohou být odswapována, takže pokud se dělají podobné testy, tak je vhodné před testem swap vypnot.
Ono je otazka, zda dnes naopak vubec jeste swap zapinat.
10.7.2015 19:03 R
Rozbalit Rozbalit vše Re: Srovnání i5-2400@3.10GHz s Atomem C2750@2.40GHz
Presne tak, server by mal mat dostatocne mnozstvo RAM na veci, co tam bezia. To moze byt napriklad v pripade mail servera aj 1 GB. Na mnohych serveroch swap roky nepouzivam.
Jendа avatar 11.7.2015 03:32 Jendа | skóre: 78 | blog: Jenda | JO70FB
Rozbalit Rozbalit vše Re: Srovnání i5-2400@3.10GHz s Atomem C2750@2.40GHz
Ano. Leakující nebo dlouho idlící aplikace se odswapují a volná paměť se může použít pro pagecache pomalých disků.
Já to s tou denacifikací Slovenska myslel vážně.
10.7.2015 15:56 Honza Jaroš | skóre: 6 | blog: moje_strana_plotu | Bohnice
Rozbalit Rozbalit vše Re: Srovnání i5-2400@3.10GHz s Atomem C2750@2.40GHz
Zkouším to teď znova na tmpfs, na i5 jsou rozdíly maximálně o tři vteřiny, ale na Atomu "osmijádrová" kompilace dotáhla náskok a v podstatě se neliší od i5. Dokončuju ještě test na čtyřech jádrech, pak to doplním do článku.
Jezekus avatar 10.7.2015 11:50 Jezekus | skóre: 19 | blog: jezkova_nora
Rozbalit Rozbalit vše Re: Srovnání i5-2400@3.10GHz s Atomem C2750@2.40GHz
Odpovědět | Sbalit | Link | Blokovat | Admin
Díky za zajímavé srovnani sam mám C2750 v domácím serviku už rok a jsem moc spokojen. Mám Asrock C2750d4i. Spotřeba naprosto parádní i kdyz ne priliz rozdílná v režimu idle/load.
10.7.2015 20:03 Honza Jaroš | skóre: 6 | blog: moje_strana_plotu | Bohnice
Rozbalit Rozbalit vše Re: Srovnání i5-2400@3.10GHz s Atomem C2750@2.40GHz
Odpovědět | Sbalit | Link | Blokovat | Admin
Mimochodem, zkusil jsem si ten benchmark OpenSSL spustit na počítači s Xeonem X3440, který podle specifikací nemá podporu AES-NI (a nehlásí ji ani v /proc/cpuinfo). Jak je možné, že to s parametrem -evp jede víc než dvakrát tak rychle? 
~ # openssl speed aes-128-cbc
Doing aes-128 cbc for 3s on 16 size blocks: 13336477 aes-128 cbc's in 3.00s
Doing aes-128 cbc for 3s on 64 size blocks: 3619319 aes-128 cbc's in 3.00s
Doing aes-128 cbc for 3s on 256 size blocks: 908566 aes-128 cbc's in 2.97s
Doing aes-128 cbc for 3s on 1024 size blocks: 479547 aes-128 cbc's in 3.00s
Doing aes-128 cbc for 3s on 8192 size blocks: 60609 aes-128 cbc's in 3.00s
OpenSSL 1.0.1o 12 Jun 2015
built on: Sun Jun 21 15:25:42 2015
options:bn(64,64) rc4(16x,int) des(idx,cisc,16,int) aes(partial) idea(int) blowfish(idx) 
compiler: x86_64-pc-linux-gnu-gcc -I. -I.. -I../include  -fPIC -DOPENSSL_PIC -DZLIB -DOPENSSL_THREADS -D_REENTRANT -DDSO_DLFCN -DHAVE_DLFCN_H -DL_ENDIAN -Wall -DOPENSSL_IA32_SSE2 -DOPENSSL_BN_ASM_MONT -DOPENSSL_BN_ASM_MONT5 -DOPENSSL_BN_ASM_GF2m -DSHA1_ASM -DSHA256_ASM -DSHA512_ASM -DMD5_ASM -DAES_ASM -DVPAES_ASM -DBSAES_ASM -DWHIRLPOOL_ASM -DGHASH_ASM -march=nocona -O2 -pipe -fno-strict-aliasing -Wa,--noexecstack
The 'numbers' are in 1000s of bytes per second processed.
type             16 bytes     64 bytes    256 bytes   1024 bytes   8192 bytes
aes-128 cbc      71127.88k    77212.14k    78314.11k   163685.38k   165502.98k

~ # openssl speed -evp aes-128-cbc
Doing aes-128-cbc for 3s on 16 size blocks: 38902595 aes-128-cbc's in 3.00s
Doing aes-128-cbc for 3s on 64 size blocks: 11123415 aes-128-cbc's in 3.00s
Doing aes-128-cbc for 3s on 256 size blocks: 2852978 aes-128-cbc's in 3.00s
Doing aes-128-cbc for 3s on 1024 size blocks: 710868 aes-128-cbc's in 3.00s
Doing aes-128-cbc for 3s on 8192 size blocks: 88929 aes-128-cbc's in 3.00s
OpenSSL 1.0.1o 12 Jun 2015
built on: Sun Jun 21 15:25:42 2015
options:bn(64,64) rc4(16x,int) des(idx,cisc,16,int) aes(partial) idea(int) blowfish(idx) 
compiler: x86_64-pc-linux-gnu-gcc -I. -I.. -I../include  -fPIC -DOPENSSL_PIC -DZLIB -DOPENSSL_THREADS -D_REENTRANT -DDSO_DLFCN -DHAVE_DLFCN_H -DL_ENDIAN -Wall -DOPENSSL_IA32_SSE2 -DOPENSSL_BN_ASM_MONT -DOPENSSL_BN_ASM_MONT5 -DOPENSSL_BN_ASM_GF2m -DSHA1_ASM -DSHA256_ASM -DSHA512_ASM -DMD5_ASM -DAES_ASM -DVPAES_ASM -DBSAES_ASM -DWHIRLPOOL_ASM -DGHASH_ASM -march=nocona -O2 -pipe -fno-strict-aliasing -Wa,--noexecstack
The 'numbers' are in 1000s of bytes per second processed.
type             16 bytes     64 bytes    256 bytes   1024 bytes   8192 bytes
aes-128-cbc     207480.51k   237299.52k   243454.12k   242642.94k   242835.46k
10.7.2015 20:13 Honza Jaroš | skóre: 6 | blog: moje_strana_plotu | Bohnice
Rozbalit Rozbalit vše Re: Srovnání i5-2400@3.10GHz s Atomem C2750@2.40GHz
Aha, zpět, já měl nějak zafixováno, že -evp zapíná AES-NI, ale ve skutečnosti to zapíná nějaké další kryptografické funkce (které umí AES-NI využít, ale fungují i bez něj).
pushkin avatar 11.7.2015 22:37 pushkin | skóre: 43 | blog: FluxBlog
Rozbalit Rozbalit vše Re: Srovnání i5-2400@3.10GHz s Atomem C2750@2.40GHz
Odpovědět | Sbalit | Link | Blokovat | Admin
Pěkné srovnání, pro běžného uživatele je ale trochu výhodnější tohle: Intel Core i5-2400@3.10GHz vs Atom C2750@2.40GHz. Od pohledu je jasné, že tyhle procesory jsou neporovnatelné...
🇺🇦 Pomoc pro obranu Ukrajiny | SOS Ukrajina | Web4Ukrajina | Web4Ukraine 🇺🇦
11.7.2015 23:28 Honza Jaroš | skóre: 6 | blog: moje_strana_plotu | Bohnice
Rozbalit Rozbalit vše Re: Srovnání i5-2400@3.10GHz s Atomem C2750@2.40GHz
Co si představujete pod pojmem "neporovnatelné"? :-)
xkucf03 avatar 12.7.2015 17:08 xkucf03 | skóre: 49 | blog: xkucf03
Rozbalit Rozbalit vše Re: Srovnání i5-2400@3.10GHz s Atomem C2750@2.40GHz

Rok 2010 vs. 2014, TDP 95 W vs. 20 W, 4 jádra vs. 8 jader…

Porovnat se dá samozřejmě všechno, ale u příliš odlišných věcí nedává výsledek moc velký smysl.

Mám rád, když se lidé přou, znamená to, že vědí, co dělají, a že mají směr. Frantovo.cz, SQL-DK, Relational pipes
12.7.2015 21:12 Honza Jaroš | skóre: 6 | blog: moje_strana_plotu | Bohnice
Rozbalit Rozbalit vše Re: Srovnání i5-2400@3.10GHz s Atomem C2750@2.40GHz

Rok 2010 vs. 2014, TDP 95 W vs. 20 W, 4 jádra vs. 8 jader…

Porovnat se dá samozřejmě všechno, ale u příliš odlišných věcí nedává výsledek moc velký smysl.

Jak jsem psal, srovnával jsem si to z čistě praktického důvodu, protože prostě uvažuju o nahrazení jednoho druhým. Takže pro mě to smysl má...
pushkin avatar 14.7.2015 09:48 pushkin | skóre: 43 | blog: FluxBlog
Rozbalit Rozbalit vše Re: Srovnání i5-2400@3.10GHz s Atomem C2750@2.40GHz
Jenze od zacatku je jasne, ze to bude vykonostne krok dolu. Pokud porovnavat, tak bych k tomu pridal procesor z "Atomoveho" segmentu: Athlon 5350. Pri TDP 25 W ma oproti Atomu vyssi vykon jednoho threadu, takze pokud nedojde k vyrazne paralelizaci, bude se zdat rychlejsi. Navic ma dve vyhody - integrovanou grafiku a to, ze jde o samostatny modul, neni tedy problem jej vymenit, pripadne jej zasadit do jine desky.
🇺🇦 Pomoc pro obranu Ukrajiny | SOS Ukrajina | Web4Ukrajina | Web4Ukraine 🇺🇦
14.7.2015 10:29 Honza Jaroš | skóre: 6 | blog: moje_strana_plotu | Bohnice
Rozbalit Rozbalit vše Re: Srovnání i5-2400@3.10GHz s Atomem C2750@2.40GHz
Panebože...

Já si to kvůli vám fakt kupovat nebudu. Já potřebuju úspornou serverovou desku s IPMI a rozumným výkonem ve více vláknech (který je s tím i5 srovnatelný, což mě mile překvapilo). To mi tahle věc splňuje. Dělal jsem si srovnání pro sebe a když už jsem ho měl, tak jsem ho vyblil i sem. Kdybych věděl, že kolem toho bude taková kovbojka, vyprdnul bych se na to...
14.7.2015 10:30 Honza Jaroš | skóre: 6 | blog: moje_strana_plotu | Bohnice
Rozbalit Rozbalit vše Re: Srovnání i5-2400@3.10GHz s Atomem C2750@2.40GHz
PS: ...a integrovaná grafika je mi v domácím serveru na dvě věci...
pavlix avatar 14.7.2015 11:48 pavlix | skóre: 54 | blog: pavlix
Rozbalit Rozbalit vše Re: Srovnání i5-2400@3.10GHz s Atomem C2750@2.40GHz
To je tak když lidi řeší víc čísla než užití dané věci. Já jsem teď taky koupil na jeden účel výrazně slabší hardware s ohledem na jiné výhody.
Já už tu vlastně ani nejsem. Abclinuxu umřelo.
14.7.2015 14:09 ::: | skóre: 14 | blog: e_lama
Rozbalit Rozbalit vše Re: Srovnání i5-2400@3.10GHz s Atomem C2750@2.40GHz
Chces aby to bylo usporne a integrovana grafika je ti k nicemu ? ;-)
14.7.2015 15:28 Honza Jaroš | skóre: 6 | blog: moje_strana_plotu | Bohnice
Rozbalit Rozbalit vše Re: Srovnání i5-2400@3.10GHz s Atomem C2750@2.40GHz
A grafika je mi na serveru k čemu, že jsem tak smělý? I bez toho základního VGAčka bych se obešel, když tam budu mít k dispozici BMC s možností připojit se na sériovou konzoli.
pushkin avatar 14.7.2015 14:11 pushkin | skóre: 43 | blog: FluxBlog
Rozbalit Rozbalit vše Re: Srovnání i5-2400@3.10GHz s Atomem C2750@2.40GHz
Vsak ja take nechci, aby jsi si to kupoval :-) To byla spise ma reakce na to, co si predstavuji pod pojmem (ne)porovnatelne. :-)
🇺🇦 Pomoc pro obranu Ukrajiny | SOS Ukrajina | Web4Ukrajina | Web4Ukraine 🇺🇦
14.7.2015 15:24 Honza Jaroš | skóre: 6 | blog: moje_strana_plotu | Bohnice
Rozbalit Rozbalit vše Re: Srovnání i5-2400@3.10GHz s Atomem C2750@2.40GHz
OK.
14.7.2015 22:30 Miriam | skóre: 3 | blog: zivot
Rozbalit Rozbalit vše Re: Srovnání i5-2400@3.10GHz s Atomem C2750@2.40GHz
Jo, jsou zdejší uživatelé ukňouraní čůráci. Ještě štěstí, že tu není kyknos.

Založit nové vlákno • Nahoru

ISSN 1214-1267, (c) 1999-2007 Stickfish s.r.o.