东北大学学报(自然科学版) ›› 2023, Vol. 44 ›› Issue (2): 168-176.DOI: 10.12068/j.issn.1005-3026.2023.02.003

• 信息与控制 • 上一篇    下一篇

基于改进蚁群算法的天基资源调度研究与仿真

耿蓉, 张昭, 牛天水, 王宇飞   

  1. (东北大学 计算机科学与工程学院, 辽宁 沈阳110169)
  • 修回日期:2021-12-16 接受日期:2021-12-16 发布日期:2023-02-27
  • 通讯作者: 耿蓉
  • 作者简介:耿蓉(1979-),女,辽宁沈阳人,东北大学副教授.
  • 基金资助:
    中央高校基本科研业务费专项资金资助项目(N2116015,N2116020); 辽宁省医工交叉基金资助项目(2021-YGJC-24).

Research and Simulation of Space-based Resource Scheduling Based on Improved Ant Colony Algorithm

GENG Rong, ZHANG Zhao, NIU Tian-shui, WANG Yu-fei   

  1. School of Computer Science & Engineering, Northeastern University, Shenyang 110169, China.
  • Revised:2021-12-16 Accepted:2021-12-16 Published:2023-02-27
  • Contact: WANG Yu-fei
  • About author:-
  • Supported by:
    -

摘要: 天基信息网中卫星资源有限,在轨升级难度大,链路间通信时延高,导致大规模并发任务处理效率低下.针对任务简单并发且每个任务由一个节点处理的情况,构建基于动态优先级的任务模型,对天基信息网计算与存储资源构建基于模糊聚类理论的资源模型.提出基于改进蚁群算法的天基资源调度策略,引入负载均衡因子,改变信息素更新规则,调整任务分配策略,结合Min-Min算法促进任务执行及资源分配.仿真结果表明,本文算法和对比算法相比,任务完成时间缩短29.2%,任务累积价值高出37.9%,资源负载均衡度缩小75.5%,资源利用率高出22.4%,验证了本文算法的优异性.

关键词: 天基信息网;任务动态排序;资源聚类;蚁群算法;资源调度

Abstract: The satellite resources of the space-based information network are limited.It is difficult to upgrade in orbit and the delay of inter-link communication is high, which results in inefficient processing of large-scale concurrent tasks. A dynamic priority-based task model is constructed for situations where the task is simple and concurrent and each task is handled by one node. A resource model based on fuzzy clustering theory is constructed for computing and storage resources in space-based information network. Space-based resource scheduling strategy based on improved ant colony algorithm is proposed. Load balancing factor is introduced. Pheromone update rule is changed. Task allocation strategy is adjusted. Min-Min-algorithm is combined to promote task execution and resource allocation. Simulation results show that compared with the comparison algorithm, the task completion time is 29.2% shorter, the task cumulative value is 37.9% higher, the resource load equilibrium degree is 75.5% smaller, and the resource utilization rate is 22.4% higher, which verifies the excellence of the algorithm.

Key words: space-based information network; task dynamic sorting; resource clustering; ant colony algorithm; resource scheduling

中图分类号: