作者投稿和查稿 主编审稿 专家审稿 编委审稿 远程编辑

计算机工程

• 体系结构与软件技术 • 上一篇    下一篇

一种环境感知的仿真服务调度模型

张学军,闫光辉,胡晓辉   

  1. (兰州交通大学电子与信息工程学院,兰州 730070)
  • 收稿日期:2013-05-29 出版日期:2014-03-15 发布日期:2014-03-13
  • 作者简介:张学军(1977-),男,讲师、博士研究生,主研方向:服务计算,服务仿真;闫光辉、胡晓辉,教授。
  • 基金资助:
    国家自然科学基金资助项目(61163009, 61163010)。

A Context-aware Simulation Service Scheduling Model

ZHANG Xue-jun, YAN Guang-hui, HU Xiao-hui   

  1. (School of Electronics and Information Engineering, Lanzhou Jiaotong University, Lanzhou 730070, China)
  • Received:2013-05-29 Online:2014-03-15 Published:2014-03-13

摘要: 针对传统分布式仿真框架HLA/RTI中动态调度负载平衡较差、执行可靠性较低的问题,提出一种基于环境感知的仿真服务调度模型C3SM。C3SM模型包含总体结构、调度策略和服务部署。总体结构规定各个模块的功能与交互接口,调度策略采用改进的蚁群算法进行仿真服务的动态调度,服务部署采用重叠覆盖部署策略以保证较好的服务可用性和较低的资源消耗。仿真实验结果表明,与传统分布式仿真框架HLA/RTI相比,C3SM模型通过实时获取执行环境的上下文信息,能够获得较优的负载平衡度,且重叠覆盖的仿真服务部署策略使得仿真运行系统的可靠性明显提高。

关键词: 环境感知, 高层体系结构, 分布式仿真, Web 服务, 蚁群算法, 仿真服务调度

Abstract: In order to address the problem of traditional distribution simulation framework HLA/RTI with respect to poor load balancing performance in service dynamic scheduling and low reliability of service execution, this paper proposes a Context-aware Simulation Service Scheduling Model(C3SM), which includes general framework, scheduling strategy and service deployment. The framework provides the functions and interaction interfaces of each module. For the scheduling scheme, a modified ant colony algorithm is used to achieve optimum load balancing and system reliability. Moreover, the overlapped coverage deployment strategy is adopted to obtain the high service availability and low resource consumption in the service deployment. Experiments are carried out on performance comparisons between the traditional HLA/RTI and C3SM in the scheduling strategy and the reliability of service performing, the results show that C3SM can obtain good load balance with the real-time context information of the execution environment, and the overlapped coverage simulation service deployment scheme greatly improves the reliability of the simulation execution system.

Key words: context-aware, High Level Architecture(HLA), distributed simulation, Web service, ant colony algorithm, simulation service scheduling

中图分类号: