Optimization Technology Of Cryptographic Algorithms For Multi-core Platform

As an important means of guaranteeing information security,cryptographic algorithms have been widely used in various fields.The ever-increasing demand for secure applications in various fields puts forward the never-ending requirements for improving the performance of cryptographic algorithms.With the emergence of various high-performance computing devices,it is very important to study the performance optimization techniques of architecture-based cryptographic processing.In this paper,parallel optimization techniques for typical cryptographic algorithms are studied on the basis of high performance multi-core platform.For DES and MD5,a high throughput fine granularity optimization algorithm based on GPU and MIC is proposed.The influence of different optimized granularity on the performance of the algorithm is discussed.The experimental results show that the parallel optimization method based on appropriate granularity can obtain high throughput performance.For the password cracking application that can reflect the password processing performance,this paper proposes a parallel optimization technique for cracked symmetric ciphers using PDFcrack.load balancing high-throughput multi-thread parallel processing algorithm based on MIC platform is applied in it.It gets 30 times better speed ratio than the mainstream general-purpose microprocessor platform.A kernel parallel optimization method for Block Wiedemann(BW)algorithm for asymmetric cryptography is proposed.It applies a Multithreading parallel processing algorithm based on large number of blocks and hidden by GPU platform and gets 7.44 times better speed ratio than the mainstream general-purpose microprocessor platform.