.gpu pull away ur cpu.

.so bring it on, gpu “graphics processing unit” istilah yang cukup sering di dengar dalam industri graphic card dikarenakan vga sekarang ini,menerapkan proses komputasi secara terpisah dengan cpu. nvidia dengan cuda-nya,ati dengan ati stream-nya dan dalam perkembangan gpu menerapkan sistem multi core. 240 core [at] nvidia gt 280, cpu sendiri hanya 2/4 core yang umum beredar.

.walau dalam tujuan komersial, pihak rusia mulai menerapkan gpu dalam proses cracking. dari kubu open source ada beberapa project yang menarik dari pyrit sampai CUDA-Multiforcer. dimana juga menerapkan gpu dalam proses crackingnya. gpu cracking lebih menitik beratkan ke cuda dimana dukungan driver dari nvidia lebih baik.
.let’s join our class.

.in wind#ws with BarsWF MD5.
.barsWF melakukan proses bruteforce terhadap single hash md5 dengan bantuan gpu. akhir-akhir ini bahkan sering digunakan sebagai ajang benchmark sebuah mesin.

.–with gpu–.

+-----------------------------------------------------------+
? BarsWF MD5 bruteforcer v0.8 ?       http://3.14.by/en/md5 ?
? by Svarychevski Michail             http://3.14.by/ru/md5 ?
?-----------------------------------------------------------?
?    GPU0:  243.92 MHash/sec     CPU0:   42.92 MHash/sec    ?
?                                CPU1:   36.80 MHash/sec    ?
?                                                           ?
?    GPU*:  243.92 MHash/sec     CPU*:   79.72 MHash/sec    ?
?-----------------------------------------------------------?
? Key: %?u*q                Avg.Total:  314.76 MHash/sec    ?
? Hash:2E86E42F251DC797E33C41674AF40053                     ?
? Progress: 90.84 % ETC    0 days  0 hours  0 min  1 sec    ?
+-----------------------------------------------------------+
-- Key is: d0tZ-??-------------------------------------------
found at 12second

.–doesn’t with gpu–.

+-----------------------------------------------------------+
? BarsWF MD5 bruteforcer v0.8 ?       http://3.14.by/en/md5 ?
? by Svarychevski Michail             http://3.14.by/ru/md5 ?
?-----------------------------------------------------------?
?                                CPU0:   41.65 MHash/sec    ?
?                                CPU1:   42.87 MHash/sec    ?
?                                                           ?
?    GPU*:    0.00 MHash/sec     CPU*:   84.52 MHash/sec    ?
?-----------------------------------------------------------?
? Key: U(tjt                Avg.Total:   80.38 MHash/sec    ?
? Hash:2E86E42F251DC797E33C41674AF40053                     ?
? Progress: 90.57 % ETC    0 days  0 hours  0 min  5 sec    ?
+-----------------------------------------------------------+
-- Key is: d0tZ-??-------------------------------------------
found at 54second

.seperti yang dapat dilihat, penggunaan tanpa proses gpu memakan waktu 4.5 kalinya daripada proses pertama.

$ lspci | grep -i vga
01:00.0 VGA compatible controller: nVidia Corporation G94 [GeForce 9600 GT] (rev a1)

$ lspci -v -s 01:00.0
01:00.0 VGA compatible controller: nVidia Corporation G94 [GeForce 9600 GT] (rev a1)
Flags: bus master, fast devsel, latency 0, IRQ 16
Memory at fa000000 (32-bit, non-prefetchable) [size=16M]
Memory at c0000000 (64-bit, prefetchable) [size=512M]
Memory at f8000000 (64-bit, non-prefetchable) [size=32M]
I/O ports at af00 [size=128]
[virtual] Expansion ROM at fb000000 [disabled] [size=512K]
Capabilities: [60] Power Management version 3
Capabilities: [68] Message Signalled Interrupts: Mask- 64bit+ Queue=0/0 Enable-
Capabilities: [78] Express Endpoint, MSI 00
Capabilities: [b4] Vendor Specific Information
Capabilities: [100] Virtual Channel
Capabilities: [128] Power Budgeting
Capabilities: [600] Vendor Specific Information
Kernel driver in use: nvidia
Kernel modules: nvidia, nvidiafb

.in linux with CUDA-Multiforcer.
.mensupport hash md4,md5 dan ntlm, bahkan proses bruteforce lebih cepat untuk huge hashlist. daripada BarsWF yang berjalan pada paltform wind#ws, blame micros#ft.

$ ./CUDA-Multiforcer -c charsets/charsetfull -f
test_hash_files/testhash.txt --min 5 --max 5 -h MD5
Launching 64-bit binary
Cryptohaze.com CUDA Multiforcer (multiple hash brute forcer)
by Bitweasil
Version 0.61 beta, length 0-14
Currently supported hash types: MD5 MD4 NTLM
Hash type: MD5
CUDA Device Information:
Device 0: "GeForce 9600 GT"
Number of cores:                               64
Clock rate:                                    1.60 GHz
Charset loaded (96 characters)
Hashes loaded (2 hashes)
Launching kernel for password length 5
Done: 54.44%  Step rate: 166.1M/s Search rate: 332.1M/sec
MD5 : 6BC018C635FC7251BC25D0E11984021E: 'h4X0r'    0x6834583072
Done: 94.62%  Step rate: 168.3M/s Search rate: 336.5M/sec
MD5 : 2E86E42F251DC797E33C41674AF40053: 'd0tZ-'    0x6430745A2D
Done: 98.64%  Step rate: 168.3M/s Search rate: 336.5M/sec

------------------------------------------

Compute done: Reference time 48.6 seconds
Stepping rate: 167.8M MD4/s
Search rate: 335.5M NTLM/s

.in linux with pyrit.
.wifoo freak, pyrit bergerak dalam wpa & wpa2 hash table generation. fleksibel, output bisa digunakan bersama cow patty dan airolib-ng.

$ ./pyrit benchmark
Pyrit 0.2.2 (C) 2008, 2009 Lukas Lueg http://pyrit.googlecode.com
This code is distributed under the GNU General Public License v3

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

Running benchmark for at least 60 seconds...

CUDA-Device #1 'GeForce 9600 GT': 3176.95 PMKs/s, 98.14% occupancy
CPU-Core (x86): 259.89 PMKs/s, 98.33% occupancy

Benchmark done. 3436.84 PMKs/s total.

.*tested [on] [email protected],4 gb ram and nvidia 9600gt.

.bila diterapkan secara distribusi seperti pada ezine sebelumnya.
.keeewwlll house of cracking will be born.

.gamer freak or hash freak, once again make ur choice,

.EOF–

2 Responses to “.gpu pull away ur cpu.”

1. df0x says:

   March 28, 2009 at 10:27 am

   GPU oh GPU, mengingatkan saya pada seorang direktur sebuah PT di Bandung dengan dagangan Apple-nya 🙂

   Reply
2. poniman_coy says:

   March 29, 2009 at 10:32 am

   wah enak jadi gamer aja deh…dr pada liat md5…mending liat game.fire in the hole…wjwjw

   Reply