论文部分内容阅读
为了对任意长的明/密文进行并行加密/解密,在分析了AES(Advanced Encryption Standard)的多种模式后,采用了最合适的CTR(Counter)模式设计和实现了GPU(Graphics Processing Unit)并行AES算法,并进行了优化.针对Nvidia Geforce GTX460平台理论分析了CTR模式的并行AES算法和串行AES算法的时间复杂度,得到综合的加速比为31.59,然后在Nvidia Geforce GTX 460平台上实验运行,结果显示CTR模式的AES-256的GPU并行算法相对串行CTR模式的AES算法,实验加速比跟理论加速比基本吻合.在此基础上,对CTR模式的AES-256进行了优化.实验结果显示,优化的CTR模式的AES-256并行算法在加速比上随着明文的增大提升的比例渐渐减少并趋于稳定.优化的CTR模式的AES算法加密数据量小的明文时,其优化效果更为明显,故能有效地提升SSL(Secure Socket Layer,其明文区间为35KB-150KB)的加密性能.
In order to encrypt / decrypt any arbitrary long / cipher text in parallel, the most suitable CTR (Counter) mode is designed and implemented after analyzing various modes of AES (Advanced Encryption Standard) and a GPU (Graphics Processing Unit) Parallel AES algorithm was optimized and optimized.According to Nvidia Geforce GTX460 platform theoretical analysis of the CTR mode parallel AES algorithm and serial AES algorithm time complexity, the overall speedup was 31.59, and then on the Nvidia Geforce GTX 460 platform experiment Run, the results show that CTR mode AES-256 GPU parallel algorithm relative to the serial CTR mode of the AES algorithm, the experimental speedup ratio and theoretical speedup basically consistent.On the basis of the CTR mode AES-256 was optimized.Experiment The results show that the proportion of AES-256 parallel algorithm with optimized CTR mode decreases and tends to be stable with the increase of plaintext in the acceleration ratio.The optimized AES algorithm of CTR mode, when encrypting plaintext with small amount of data, The effect is more obvious, so it can effectively improve the SSL (Secure Socket Layer, the plaintext interval of 35KB-150KB) encryption performance.