Nvidia Grid K2 card performance = bad
#1
Hi all,

Just posting benchmark for an Nvidia Grid K2 card. Performance is much less that what I was expecting, considering it has 3072 cuda cores.  I am using vmware and PCI passthru mode, which might have something to do with it.


Device #1: GRID K2, 4095MB, 745Mhz, 8MCU
Device #2: GRID K2, 4095MB, 745Mhz, 8MCU

Cuda Cores:  3072
Nvidia Forward driver: 352.68


Code:
Hashtype: MD4 Workload: 1024 loops, 256 accel Speed.GPU.#1.:  3460.1 MH/s Speed.GPU.#2.:  3460.6 MH/s Speed.GPU.#*.:  6920.7 MH/s Hashtype: MD5 Workload: 1024 loops, 256 accel Speed.GPU.#1.:  2462.3 MH/s Speed.GPU.#2.:  2461.5 MH/s Speed.GPU.#*.:  4923.8 MH/s Hashtype: Half MD5 Workload: 1024 loops, 256 accel Speed.GPU.#1.:   725.7 MH/s Speed.GPU.#2.:   725.8 MH/s Speed.GPU.#*.:  1451.5 MH/s Hashtype: SHA1 Workload: 1024 loops, 256 accel Speed.GPU.#1.:   653.8 MH/s Speed.GPU.#2.:   653.7 MH/s Speed.GPU.#*.:  1307.4 MH/s Hashtype: SHA256 Workload: 1024 loops, 256 accel Speed.GPU.#1.:   267.8 MH/s Speed.GPU.#2.:   267.8 MH/s Speed.GPU.#*.:   535.6 MH/s Hashtype: SHA384 Workload: 256 loops, 256 accel Speed.GPU.#1.: 67878.6 kH/s Speed.GPU.#2.: 67881.6 kH/s Speed.GPU.#*.:   135.8 MH/s Hashtype: SHA512 Workload: 256 loops, 256 accel Speed.GPU.#1.: 67832.4 kH/s Speed.GPU.#2.: 67827.6 kH/s Speed.GPU.#*.:   135.7 MH/s Hashtype: SHA-3(Keccak) Workload: 128 loops, 256 accel Speed.GPU.#1.: 58367.9 kH/s Speed.GPU.#2.: 58372.3 kH/s Speed.GPU.#*.:   116.7 MH/s Hashtype: SipHash Workload: 1024 loops, 256 accel Speed.GPU.#1.:  3057.6 MH/s Speed.GPU.#2.:  3057.6 MH/s Speed.GPU.#*.:  6115.2 MH/s Hashtype: RipeMD160 Workload: 1024 loops, 256 accel Speed.GPU.#1.:   488.4 MH/s Speed.GPU.#2.:   488.4 MH/s Speed.GPU.#*.:   976.8 MH/s Hashtype: Whirlpool Workload: 512 loops, 32 accel Speed.GPU.#1.: 41075.1 kH/s Speed.GPU.#2.: 41046.9 kH/s Speed.GPU.#*.: 82122.0 kH/s Hashtype: GOST R 34.11-94 Workload: 512 loops, 64 accel Speed.GPU.#1.: 39296.4 kH/s Speed.GPU.#2.: 39201.0 kH/s Speed.GPU.#*.: 78497.3 kH/s Hashtype: GOST R 34.11-2012 (Streebog) 256-bit Workload: 512 loops, 16 accel Speed.GPU.#1.:  9176.3 kH/s Speed.GPU.#2.:  9175.1 kH/s Speed.GPU.#*.: 18351.4 kH/s Hashtype: GOST R 34.11-2012 (Streebog) 512-bit Workload: 512 loops, 16 accel Speed.GPU.#1.:  9159.0 kH/s Speed.GPU.#2.:  9187.7 kH/s Speed.GPU.#*.: 18346.7 kH/s Hashtype: phpass, MD5(Wordpress), MD5(phpBB3), MD5(Joomla) Workload: 1024 loops, 32 accel Speed.GPU.#1.:   604.3 kH/s Speed.GPU.#2.:   603.1 kH/s Speed.GPU.#*.:  1207.4 kH/s Hashtype: scrypt Workload: 1 loops, 64 accel Speed.GPU.#1.:    23690 H/s Speed.GPU.#2.:    23694 H/s Speed.GPU.#*.:    47384 H/s Hashtype: PBKDF2-HMAC-MD5 Workload: 1000 loops, 8 accel Speed.GPU.#1.:   662.7 kH/s Speed.GPU.#2.:   662.6 kH/s Speed.GPU.#*.:  1325.2 kH/s Hashtype: PBKDF2-HMAC-SHA1 Workload: 1000 loops, 8 accel Speed.GPU.#1.:   309.6 kH/s Speed.GPU.#2.:   309.6 kH/s Speed.GPU.#*.:   619.3 kH/s Hashtype: PBKDF2-HMAC-SHA256 Workload: 1000 loops, 8 accel Speed.GPU.#1.:   107.4 kH/s Speed.GPU.#2.:   107.3 kH/s Speed.GPU.#*.:   214.7 kH/s Hashtype: PBKDF2-HMAC-SHA512 Workload: 1000 loops, 8 accel Speed.GPU.#1.:    30230 H/s Speed.GPU.#2.:    30207 H/s Speed.GPU.#*.:    60437 H/s Hashtype: Skype Workload: 1024 loops, 256 accel Speed.GPU.#1.:   737.8 MH/s Speed.GPU.#2.:   737.7 MH/s Speed.GPU.#*.:  1475.5 MH/s Hashtype: WPA/WPA2 Workload: 1024 loops, 32 accel Speed.GPU.#1.:    38806 H/s Speed.GPU.#2.:    38816 H/s Speed.GPU.#*.:    77621 H/s Hashtype: IKE-PSK MD5 Workload: 512 loops, 128 accel Speed.GPU.#1.:   194.4 MH/s Speed.GPU.#2.:   194.5 MH/s Speed.GPU.#*.:   388.9 MH/s Hashtype: IKE-PSK SHA1 Workload: 512 loops, 128 accel Speed.GPU.#1.: 70734.3 kH/s Speed.GPU.#2.: 70835.6 kH/s Speed.GPU.#*.:   141.6 MH/s Hashtype: NetNTLMv1-VANILLA / NetNTLMv1+ESS Workload: 1024 loops, 256 accel Speed.GPU.#1.:  1636.0 MH/s Speed.GPU.#2.:  1636.6 MH/s Speed.GPU.#*.:  3272.6 MH/s Hashtype: NetNTLMv2 Workload: 1024 loops, 32 accel Speed.GPU.#1.:   158.9 MH/s Speed.GPU.#2.:   158.7 MH/s Speed.GPU.#*.:   317.6 MH/s Hashtype: IPMI2 RAKP HMAC-SHA1 Workload: 256 loops, 256 accel Speed.GPU.#1.:   161.0 MH/s Speed.GPU.#2.:   161.0 MH/s Speed.GPU.#*.:   321.9 MH/s Hashtype: Kerberos 5 AS-REQ Pre-Auth etype 23 Workload: 256 loops, 32 accel Speed.GPU.#1.:  5845.2 kH/s Speed.GPU.#2.:  5845.3 kH/s Speed.GPU.#*.: 11690.5 kH/s Hashtype: DNSSEC (NSEC3) Workload: 512 loops, 256 accel Speed.GPU.#1.:   314.8 MH/s Speed.GPU.#2.:   314.8 MH/s Speed.GPU.#*.:   629.6 MH/s Hashtype: PostgreSQL Challenge-Response Authentication (MD5) Workload: 1024 loops, 256 accel Speed.GPU.#1.:   677.3 MH/s Speed.GPU.#2.:   677.5 MH/s Speed.GPU.#*.:  1354.9 MH/s Hashtype: MySQL Challenge-Response Authentication (SHA1) Workload: 1024 loops, 256 accel Speed.GPU.#1.:   223.8 MH/s Speed.GPU.#2.:   223.7 MH/s Speed.GPU.#*.:   447.5 MH/s Hashtype: SIP digest authentication (MD5) Workload: 1024 loops, 32 accel Speed.GPU.#1.:   415.9 MH/s Speed.GPU.#2.:   415.9 MH/s Speed.GPU.#*.:   831.9 MH/s Hashtype: SMF > v1.1 Workload: 512 loops, 256 accel Speed.GPU.#1.:   263.7 MH/s Speed.GPU.#2.:   263.9 MH/s Speed.GPU.#*.:   527.6 MH/s Hashtype: vBulletin < v3.8.5 Workload: 1024 loops, 256 accel Speed.GPU.#1.:   665.1 MH/s Speed.GPU.#2.:   665.8 MH/s Speed.GPU.#*.:  1330.9 MH/s Hashtype: vBulletin > v3.8.5 Workload: 1024 loops, 256 accel Speed.GPU.#1.:   479.5 MH/s Speed.GPU.#2.:   479.7 MH/s Speed.GPU.#*.:   959.2 MH/s Hashtype: IPB2+, MyBB1.2+ Workload: 1024 loops, 256 accel Speed.GPU.#1.:   493.4 MH/s Speed.GPU.#2.:   494.0 MH/s Speed.GPU.#*.:   987.4 MH/s Hashtype: WBB3, Woltlab Burning Board 3 Workload: 256 loops, 256 accel Speed.GPU.#1.:   128.4 MH/s Speed.GPU.#2.:   128.4 MH/s Speed.GPU.#*.:   256.8 MH/s Hashtype: Joomla < 2.5.18 Workload: 1024 loops, 256 accel Speed.GPU.#1.:  2459.9 MH/s Speed.GPU.#2.:  2459.9 MH/s Speed.GPU.#*.:  4919.9 MH/s Hashtype: PHPS Workload: 1024 loops, 256 accel Speed.GPU.#1.:   666.5 MH/s Speed.GPU.#2.:   666.5 MH/s Speed.GPU.#*.:  1333.0 MH/s Hashtype: Drupal7 Workload: 1024 loops, 8 accel Speed.GPU.#1.:     4022 H/s Speed.GPU.#2.:     4022 H/s Speed.GPU.#*.:     8044 H/s Hashtype: osCommerce, xt:Commerce Workload: 1024 loops, 256 accel Speed.GPU.#1.:   737.4 MH/s Speed.GPU.#2.:   737.4 MH/s Speed.GPU.#*.:  1474.8 MH/s Hashtype: PrestaShop Workload: 1024 loops, 256 accel Speed.GPU.#1.:   874.1 MH/s Speed.GPU.#2.:   874.1 MH/s Speed.GPU.#*.:  1748.2 MH/s Hashtype: Django (SHA-1) Workload: 512 loops, 256 accel Speed.GPU.#1.:   263.8 MH/s Speed.GPU.#2.:   263.8 MH/s Speed.GPU.#*.:   527.5 MH/s Hashtype: Django (PBKDF2-SHA256) Workload: 1024 loops, 8 accel Speed.GPU.#1.:     5430 H/s Speed.GPU.#2.:     5420 H/s Speed.GPU.#*.:    10850 H/s Hashtype: Mediawiki B type Workload: 1024 loops, 256 accel Speed.GPU.#1.:   416.1 MH/s Speed.GPU.#2.:   412.8 MH/s Speed.GPU.#*.:   828.8 MH/s Hashtype: Redmine Project Management Web App Workload: 1024 loops, 256 accel Speed.GPU.#1.:   216.4 MH/s Speed.GPU.#2.:   216.4 MH/s Speed.GPU.#*.:   432.8 MH/s Hashtype: PostgreSQL Workload: 1024 loops, 256 accel Speed.GPU.#1.:  2460.4 MH/s Speed.GPU.#2.:  2460.5 MH/s Speed.GPU.#*.:  4920.8 MH/s Hashtype: MSSQL(2000) Workload: 512 loops, 256 accel Speed.GPU.#1.:   628.5 MH/s Speed.GPU.#2.:   628.4 MH/s Speed.GPU.#*.:  1256.9 MH/s Hashtype: MSSQL(2005) Workload: 512 loops, 256 accel Speed.GPU.#1.:   627.9 MH/s Speed.GPU.#2.:   627.8 MH/s Speed.GPU.#*.:  1255.7 MH/s Hashtype: MSSQL(2012) Workload: 256 loops, 256 accel Speed.GPU.#1.: 67575.9 kH/s Speed.GPU.#2.: 67592.6 kH/s Speed.GPU.#*.:   135.2 MH/s Hashtype: MySQL323 Workload: 512 loops, 256 accel Speed.GPU.#1.:  8384.0 MH/s Speed.GPU.#2.:  8384.8 MH/s Speed.GPU.#*.: 16768.8 MH/s Hashtype: MySQL4.1/MySQL5 Workload: 512 loops, 256 accel Speed.GPU.#1.:   314.6 MH/s Speed.GPU.#2.:   314.9 MH/s Speed.GPU.#*.:   629.6 MH/s Hashtype: Oracle H: Type (Oracle 7+) Workload: 512 loops, 64 accel Speed.GPU.#1.:   106.6 MH/s Speed.GPU.#2.:   106.7 MH/s Speed.GPU.#*.:   213.3 MH/s Hashtype: Oracle S: Type (Oracle 11+) Workload: 1024 loops, 256 accel Speed.GPU.#1.:   654.0 MH/s Speed.GPU.#2.:   653.9 MH/s Speed.GPU.#*.:  1307.9 MH/s Hashtype: Oracle T: Type (Oracle 12+) Workload: 1024 loops, 8 accel Speed.GPU.#1.:     7374 H/s Speed.GPU.#2.:     7354 H/s Speed.GPU.#*.:    14729 H/s Hashtype: Sybase ASE Workload: 512 loops, 32 accel Speed.GPU.#1.: 31957.0 kH/s Speed.GPU.#2.: 31879.5 kH/s Speed.GPU.#*.: 63836.5 kH/s Hashtype: EPiServer 6.x < v4 Workload: 512 loops, 256 accel Speed.GPU.#1.:   263.8 MH/s Speed.GPU.#2.:   265.5 MH/s Speed.GPU.#*.:   529.2 MH/s Hashtype: EPiServer 6.x > v4 Workload: 512 loops, 256 accel Speed.GPU.#1.:   250.5 MH/s Speed.GPU.#2.:   250.5 MH/s Speed.GPU.#*.:   501.0 MH/s Hashtype: md5apr1, MD5(APR), Apache MD5 Workload: 1000 loops, 32 accel Speed.GPU.#1.:  1102.6 kH/s Speed.GPU.#2.:  1103.1 kH/s Speed.GPU.#*.:  2205.6 kH/s Hashtype: ColdFusion 10+ Workload: 128 loops, 128 accel Speed.GPU.#1.:   176.5 MH/s Speed.GPU.#2.:   176.5 MH/s Speed.GPU.#*.:   353.0 MH/s Hashtype: hMailServer Workload: 512 loops, 256 accel Speed.GPU.#1.:   250.6 MH/s Speed.GPU.#2.:   250.6 MH/s Speed.GPU.#*.:   501.1 MH/s Hashtype: SHA-1(Base64), nsldap, Netscape LDAP SHA Workload: 1024 loops, 256 accel Speed.GPU.#1.:   653.6 MH/s Speed.GPU.#2.:   654.0 MH/s Speed.GPU.#*.:  1307.6 MH/s Hashtype: SSHA-1(Base64), nsldaps, Netscape LDAP SSHA Workload: 1024 loops, 256 accel Speed.GPU.#1.:   654.1 MH/s Speed.GPU.#2.:   653.7 MH/s Speed.GPU.#*.:  1307.7 MH/s Hashtype: SSHA-512(Base64), LDAP {SSHA512} Workload: 256 loops, 256 accel Speed.GPU.#1.: 67813.1 kH/s Speed.GPU.#2.: 67835.3 kH/s Speed.GPU.#*.:   135.6 MH/s Hashtype: LM Workload: 1024 loops, 256 accel Speed.GPU.#1.:   475.2 MH/s Speed.GPU.#2.:   475.2 MH/s Speed.GPU.#*.:   950.5 MH/s Hashtype: NTLM Workload: 1024 loops, 256 accel Speed.GPU.#1.:  3382.1 MH/s Speed.GPU.#2.:  3381.3 MH/s Speed.GPU.#*.:  6763.4 MH/s Hashtype: Domain Cached Credentials (DCC), MS Cache Workload: 1024 loops, 256 accel Speed.GPU.#1.:  1124.9 MH/s Speed.GPU.#2.:  1124.9 MH/s Speed.GPU.#*.:  2249.8 MH/s Hashtype: Domain Cached Credentials 2 (DCC2), MS Cache 2 Workload: 1024 loops, 16 accel Speed.GPU.#1.:    31742 H/s Speed.GPU.#2.:    31729 H/s Speed.GPU.#*.:    63471 H/s
#2
It's a Kepler GPU and thus lacking LOP3.LUT, your expectations shouldn't have been very high. It's not all about cores and clock rates Wink No, in this case PCI passthru is not to blame -- all pre-Maxwell Nvidia cards suck, and Tesla/Quadro cards suck harder.
#3
Wink 
lol thanks Smile ill find another use for the Grid card and go and get myself some Maxwell cards.