Optimization Implementation of SM3 Algorithm Based on 64 Rounds Grading Calculation

This electronic document analyzes the calculation process of SM3 algorithm and finds that the iterative compression process is suitable for optimizing and improving with the characteristics of hardware. Therefore, a parallel implementation strategy of running water line and computing is proposed during the iterative compression process. This strategy makes full use of the hardware's parallel features, making the SM3 algorithm the most complex and time-consuming. In the 64-wheel calculation project, it is possible to allow successively entered data to be calculated in different rounds at the same time. In the case of continuous large traffic data, the computational efficiency is greatly improved, and the delay is reduced, and there is no case where the calculation module waits for data. This allows computing resources to be used efficiently. The delay can be as low as 0 at 125 MHz clock. 008μs. in addition, the parallel mode of multiple algorithm kernel modules is used to make the algorithm kernel become a relatively independent module, which greatly increases the scalability of the performance.