Resource Scheduling Strategy for Performance Optimization Based on Heterogeneous CPU-GPU Platform

In recent years, with the development of processor architecture, heterogeneous processors including Center processing unit (CPU) and Graphics (GPU) have become mainstream. However, due to differences core, system is now facing many problems that need be solved. order solve these problems, this paper try focus on utilization efficiency core design some reasonable resource scheduling strategies. To improve performance system, proposes a combination strategy for single task multi-task multiple tasks. The consists two sub-strategies, first improves execution tasks GPU by changing thread organization structure. second focuses working state efficient develops more workload balancing schemes systems. obtains cores global information through samples. Based information, an improved ant colony algorithm used quickly obtain allocation scheme, which fully utilizes characteristics cores. experimental results show reduces time 29.13% average. case tasks, up 23.38% based combined strategy. Both strategies can make better use resources systems significantly reduce