吉林大学学报(工学版)2024,Vol.54Issue(8) :2282-2287.DOI:10.13229/j.cnki.jdxbgxb.20230340

面向嵌入式多核系统的缓存调度算法优化

Optimizationofcacheschedulingalgorithmforembedded multi-coresystem

王鹏 杨国栋
吉林大学学报(工学版)2024,Vol.54Issue(8) :2282-2287.DOI:10.13229/j.cnki.jdxbgxb.20230340

面向嵌入式多核系统的缓存调度算法优化

Optimizationofcacheschedulingalgorithmforembedded multi-coresystem

王鹏 1杨国栋2
扫码查看

作者信息

  • 1. 中国民航大学民航航空器适航审定技术重点实验室,天津 300300;中国民航大学安全科学与工程学院,天津 300300
  • 2. 中国民航大学民航航空器适航审定技术重点实验室,天津 300300;中国民航大学中欧航空工程师学院,天津 300300
  • 折叠

摘要

为解决多核系统缓存调度缓慢、性能不佳等问题,提出了面向嵌入式多核系统的缓存调度算法优化方法.通过多核系统调度模型的建立,分析其系统总资源负载平衡情况;通过扩大调度策略对系统同步比进行计算;通过整数线性规划方程使系统达到任务负载均衡、通信开销最小化、满足片上存储限制,实现嵌入式多核系统的缓存调度优化.实验结果表明:本文方法的通信开销较低,缓存调度时间在6.38~12.32 ms,说明其缓存调度性能更佳.

Abstract

In order to solve the problems of slow cache scheduling and poor performance in multi-core systems,a cache scheduling algorithm optimization method for embedded multi-core systems is proposed.By establishing a multi-core system scheduling model,analyze the overall resource load balance of the system;Calculate the synchronization ratio of the system by expanding the scheduling strategy;By using integer linear programming equations,the system achieves task load balancing,minimizes communication overhead,and satisfies on-chip storage limitations,achieving cache scheduling optimization for embedded multi-core systems.The experimental results show that the proposed method has lower communication overhead and a cache scheduling time between 6.38-12.32 ms,indicating better cache scheduling performance.

关键词

嵌入式多核处理器/通信开销/数据流图/扩大调度策略/缓存调度优化

Key words

embedded multi-core processor/communication overhead/data flow diagram/expand the scheduling strategy/cache scheduling optimization

引用本文复制引用

出版年

2024
吉林大学学报(工学版)
吉林大学

吉林大学学报(工学版)

CSTPCD北大核心
影响因子:0.792
ISSN:1671-5497
段落导航相关论文