摘要：

为提升高级加密标准(AES)的加密性能,利用显卡的通用计算能力,在统一计算设备架构(CUDA)平台上实现AES 的128 位、192 位和256 位3 个版本的GPU 并行算法,并提出优化的AES 并行算法。在考虑块内线程数量、共享存储器容量和总块数的基础上,根据分块最优值的经验数据指导AES 算法在GPU 上的最优分块。实验结果表明,与未优化的AES 并行算法相比,该算法的3 个版本在Nvidia Geforce G210 显卡上的加密速度分别提高 5. 28% ,14． 55% 和12． 53% , 而在Nvidia Geforce GTX460 显卡上的加密速度分别提高12． 48% ,15． 40% 和 15． 84% ,且能更好地对SSL 数据进行加密。

关键词: 分块, 经验数据, 并行算法, 优化, 高级加密标准, 统一计算设备架构

Abstract:

In order to enhance the efficiency of Advanced Encryption Standard ( AES ) and make use of general computing ability of Graphics Processing Unit (GPU),all the three versions of GPU parallel AES,namely 128 bit version,192 bit version and 256 bit version,are implemented on Compute Unified Device Architecture(CUDA). Then,it proposes optimization alogorithms of parallel AES with 3 versions. These alogorithms first consider threads amount in a block,shared memory size and total blocks,then use the experience data of optimal value of block size to guide AES alogorithm’s optimal block on GPU. Experimental results show that compared with unoptimized parral AES,these alogorithms can obtain encryption mean speedup by 5． 28% ,14． 55% and 12． 53% respectively on Nvidia Geforce G210 graphics card,while by 12． 48% ,15． 40% and 15． 84% on Nvidia Geforce GTX460 graphics card. In addition,these alogorithms are better at improving encrypting of Secure Socket Layer(SSL).

Key words: block;experiential data, parallel alogorithm, optimization, Advanced Encryption Standard(AES), Compute Unified Device Architecture(CUDA)

中图分类号: 

• TP301. 6