Previous [ 1] [ 2] [ 3] [ 4] [ 5] [ 6] [ 7] [ 8] [ 9] [ 10] [ 11] [ 12] [ 13]


Journal of Information Science and Engineering, Vol. 32 No. 1, pp. 97-112 (January 2016)

GPU-Based High Performance Password Recovery Technique for Hash Functions*

1School of Information Security Engineering
Shanghai Jiao Tong University
Shanghai, 200240 P.R. China
2School of Computer Science
South China Normal University
Guangzhou, 510631 P.R. China
3State Key Laboratory of Information Security
Institute of Information Engineering
Chinese Academy of Sciences
Beijing, 100093 P.R. China

Due to the development of GPGPU (General Purpose Graphic Processing Unit) technology, GPU has been applied in many computation tasks as accelerators. In this paper, a new password recovery technique for the standardized hash functions, MD5 and SHA1, are proposed by combining the optimization methods on GPU. The performance on AMD HD7970 is 2615 mc/s for SHA1 and 6877 mc/s for MD5, which is 10 times better than the original implementation. If the length of password is limited, our GPUbased technique makes it possible to recover password from hash values in a reasonable time.

Keywords: GPU, OpenCL, password recovery, hash functions, optimization

Full Text (¥þ¤åÀÉ) Retrieve PDF document (201601_06.pdf)

Received September 2, 2014; revised December 7, 2014; accepted February 9, 2015.
Communicated by Xuejia Lai.
* Weidong Qiu is supported by New Century Excellent Talents in University of Ministry of Education under Grant NCET-12-0358, Technology Innovation Research Program, Shanghai Municipal Education Commission under Grant 12ZZ019, Supporting Program of the ¡§Twelfth Five-year Plan¡¨ for Sci and Tech Research of China under Grant 2014BAK06B00. Zheng Gong is supported by the Foundation for Distinguished Young Teachers in Higher Education of Guangdong under Grant No. Yq2013051, the Project of Science and Technology New Star of Guangzhou Pearl River (2014J2200006), and the NSFC (No. 61572028) and the Natural Science Foundation of Guangdong (No.2014A030313439).
+ The corresponding author.