基于逻辑度的嵌入式软件访存压力优化方法

Method for Optimizing Memory Access Pressure of Embedded Software Based on Logic Degree

为了提高嵌入式软件访存实时性,基于逻辑度,进行嵌入式高阶链接GPU软件访存压力优化设计.通过逻辑度排序进行存储访问的分块匹配处理,建立离散稀疏分布式融合特征集,提取嵌入式高阶链接GPU软件模糊信息特征,进行嵌入式高阶链接GPU软件访问存储过程中的优化调度和适应度调整,实现嵌入式软件访存压力优化.测试结果表明,采用该方法进行访存压力优化控制时,软件的存储开销较小,提高了嵌入式软件访存的实时性.

In order to improve the real-time performance of embedded software memory access,based on the logic degree,the memory access pressure of embedded high-order link GPU software is optimized.In order to optimize the memory access pressure of embedded software,a discrete sparse distributed fusion feature set is established,and the fuzzy information features of embedded high-order link GPU software are extracted.The optimal scheduling and fitness adjustment in the storage process of embedded high-order link GPU software access are carried out.The test results show that the storage overhead of the software is small and the real-time performance of the embedded software memory access is improved.