I hope this is interesting for some people. The last week I've had access to a SuperMicro GPU server with 2x Tesla C1080 cards, 8 cores, 12 GB memory, VelociRaptor HDD.

My goal was to test the hardware and see how it stacked up to previous computed results, more specifically the Church of Wifi WPA-PSK tables. This test is done using Pyrit, as opposed to coWPAtty. But first some benchmarks:

Code:
root@bt:~/pyrit# pyrit benchmark
Pyrit 0.2.3 (C) 2008, 2009 Lukas Lueg hxxp://pyrit.googlecode.com
This code is distributed under the GNU General Public License v3

Running benchmark for at least 60 seconds...

#1: 'CUDA-Device #1 'Tesla C1060': 12146.7 PMKs/s (Occ. 95.2%; RTT 2.6)
#2: 'CUDA-Device #2 'Tesla C1060': 12155.2 PMKs/s (Occ. 95.8%; RTT 2.8)
#3: 'CPU-Core (SSE2)': 632.6 PMKs/s (Occ. 98.6%; RTT 3.0)
#4: 'CPU-Core (SSE2)': 637.1 PMKs/s (Occ. 94.1%; RTT 3.0)
#5: 'CPU-Core (SSE2)': 636.8 PMKs/s (Occ. 93.9%; RTT 3.0)
#6: 'CPU-Core (SSE2)': 637.3 PMKs/s (Occ. 98.2%; RTT 3.0)
#7: 'CPU-Core (SSE2)': 643.8 PMKs/s (Occ. 93.5%; RTT 3.0)
#8: 'CPU-Core (SSE2)': 642.0 PMKs/s (Occ. 93.1%; RTT 3.0)

Benchmark done. 28131 PMKs/s total.
Batch mode:
Code:
root@bt:/mnt/pyr/pyrit# pyrit -f "default.cow" -e "default" batch
Pyrit 0.2.3 (C) 2008, 2009 Lukas Lueg hxxp://pyrit.googlecode.com]pyrit - Project Hosting on Google Code
This code is distributed under the GNU General Public License v3

The ESSID-blobspace seems to be empty; you should create an ESSID...

Working on ESSID 'default'
Computed 769976 PMKs so far; 25940 PMKs per second; 133785 passwords buffered.
Stopped reading workunits...
Computed 991934 PMKs so far; 26131 PMKs per second; 0 passwords buffered.ed...
All done. 26229.04 PMKs/s total.
#1: 'CUDA-Device #1 'Tesla C1060': 11889.1 PMKs/s (Occ. 98.1%; RTT 2.3)
#2: 'CUDA-Device #2 'Tesla C1060': 11851.0 PMKs/s (Occ. 92.9%; RTT 2.2)
#3: 'CPU-Core (SSE2)': 624.1 PMKs/s (Occ. 92.9%; RTT 2.5)
#4: 'CPU-Core (SSE2)': 642.2 PMKs/s (Occ. 99.5%; RTT 2.5)
#5: 'CPU-Core (SSE2)': 632.6 PMKs/s (Occ. 92.8%; RTT 2.5)
#6: 'CPU-Core (SSE2)': 635.6 PMKs/s (Occ. 93.8%; RTT 2.5)
#7: 'CPU-Core (SSE2)': 607.0 PMKs/s (Occ. 94.7%; RTT 2.6)
#8: 'CPU-Core (SSE2)': 631.3 PMKs/s (Occ. 93.3%; RTT 2.5)
Batchprocessing done.
The extended run was performed using the same Churt of Wifi wordlist and essid names.

To sum it up: runtime 10h 50m, size 39GB data set, averaging 25641 PMKs/second.

hxxp://beta.ivancover.com/wiki/images/thumb/3/3b/WPA_attack_churcofwifi_set_complete.jpg/800px-WPA_attack_churcofwifi_set_complete.jpg
hxxp://beta.ivancover.com/wiki/images/e/e5/WPA_attack_supermicro_server_overview.jpg

Details are at my wiki:
hxxp://beta.ivancover.com/wiki/index.php/WPA_Attack#SuperMicro_Tesla_GPU_server