一种IaaS模式“云训练”系统设计

系统仿真学报 ›› 2015, Vol. 27 ›› Issue (5): 1095-1104.

一种IaaS模式“云训练”系统设计

陈志佳, 朱元昌, 邸彦强, 冯少冲

军械工程学院电子与光学工程系, 石家庄 050003

收稿日期:2014-04-22 修回日期:2014-07-03 出版日期:2015-05-08 发布日期:2020-09-01
作者简介:陈志佳(1986-),男,河北唐县人,博士生,研究方向为云计算、云仿真;朱元昌(1960-),男,黑龙江哈尔滨人,博士,博导,研究方向为武器系统建模与仿真;邸彦强(1973-),男,河北保定人,博士,硕导,研究方向为武器系统建模与仿真。
基金资助:
装备预研基金(9140A04030214JB34001)

Design of IaaS Mode “Cloud Training” System

Chen Zhijia, Zhu Yuanchang, Di Yanqiang, Feng Shaochong

Department of Electronic and Optics, Ordnance Engineering College, Shijiazhuang 050003, China

Received:2014-04-22 Revised:2014-07-03 Online:2015-05-08 Published:2020-09-01

摘要/Abstract

摘要： 针对网络化模拟训练真实感不强,训练效率低等问题,提出了一种IaaS (Infrastructure as a Service)模式的“云训练”。阐述了“云训练”的体系结构和运行模式,研究了“云训练”的3项核心技术。通过GPU虚拟化技术解决云环境中终端用户3D图形图像处理能力弱的问题。根据模拟训练特点,建立用户需求模型,将模糊理论引入资源调度技术中,实现虚拟资源的动态调度;结合检查点回滚、虚拟机备份和虚拟机迁移技术,实现系统高效动态容错,保证了系统的稳定性与可靠性。实验证明,“云训练”系统改善了传统模拟训练系统中资源的按需共享能力,提升了资源利用率,保证了模拟训练效果。

关键词: 模拟训练, IaaS (基础设施即服务), 云训练, GPU虚拟化, 资源动态调度, 动态容错

Abstract: There are some problems in equipment web simulating training, including poor sense of reality, low efficiency of training and high difficulty of management. To solve those problems, Infrastructure as a Service mode “cloud training” was proposed. The architecture and operation flow were explained in detail and the key technologies of “cloud training” were focused on. By GPU virtualization technology, the problem of client 3D image processing capability was solved. According to the characters of simulation training, the user requirement model was established. Fuzzy algorithm was introduced to realize resource dynamic scheduling. To assure the stability and reliability, combined with checkpoint, virtual machine backup and migration, dynamic failure tolerance algorithm was given. The results show that resource-sharing capability and simulating training effect have been improved.

Key words: simulating training, IaaS, cloud training, GPU virtualization, dynamic resource scheduling, dynamic failure tolerance

中图分类号:

TP391

陈志佳, 朱元昌, 邸彦强, 冯少冲. 一种IaaS模式“云训练”系统设计[J]. 系统仿真学报, 2015, 27(5): 1095-1104.

Chen Zhijia, Zhu Yuanchang, Di Yanqiang, Feng Shaochong. Design of IaaS Mode “Cloud Training” System[J]. Journal of System Simulation, 2015, 27(5): 1095-1104.

参考文献

[1] Sean Marston, Zhi Li, Subhajyoti Bandyopadhyay, et al. Cloud Computing-The Business Perspective[C]// The 44th Hawaii International Conference on System Sciences. Hawaii, USA: IEEE Computer Society, 2011: 176-189.
[2] Ling Qian, Zhiguo Luo, Yujian Du, et al. Cloud Computing: An Overview[J]. Lecture Notes in Computer Science (S1804-2724), 2009, 5931(1): 626-631.
[3] 邸彦强, 朱元昌, 孟宪国, 等. 基于网格技术的多用户多任务模拟训练系统[J]. 系统仿真学报, 2008, 20(3): 643-647.
[4] Meng Xianguo, Di Yanqiang, Zhu Yuanchang, et al. System Design and Implementation:a Web-based Simulation Training System in Grid Environment[J]. Journal of System Simulation (S1004-731X), 2009, 21(4): 1202-1205.
[5] 黄安祥, 冯晓文, 李劲松, 等. 基于云计算平台的航空兵训练仿真体系结构[J]. 系统仿真学报, 2011, 23(S1): 106-109.
[6] Eva Pajorová, Ladislav Hluchý.Complicated Simulation Visualization Based on Grid and Cloud Computing[C]// The 7th International Conference on Cooperative Design, Visualization & Engineering. Mallorca: Springer Berlin Heidelberg, 2010: 211-217.
[7] 冯少冲. 基于云计算的武器装备网络化模拟训练支撑技术研究[D]. 石家庄: 军械工程学院, 2011.
[8] Su Min Jang, Won Hyuk Choi, Won Young Kim.Client Rendering Method for Desktop Virtualization Services[J]. Electronics Telecommunications Research Inst (S1225-6463), 2013, 35(2): 348-351.
[9] 钱琼芬, 李春林, 张小庆. 云数据中心虚拟资源管理研究综述[J]. 计算机应用研究, 2012, 29(7): 2411-2416.
[10] Asael Dror, Hao Zhang, B.Anil Kumar, et al. Virtualized GPU in a Virtual Machine Environment [P]. United States: US 20110102443A1, 2011.
[11] Jose Duato, Francisco D. Igual, Rafael Mayo, et al. An Efficient Implementation of GPU Virtualization in High Performance Clusters[J]. Lecture Notes in Computer Science (S0302-9743), 2010, 6043(1): 385-394.
[12] Micah Dowty, Jeremy Sugerman.GPU Virtualization on VMware's Hosted I/O Architecture[J]. SIGOPS Operation Systems Review (S0163-5980), 2009, 43(3): 73-82.
[13] Sunit Parmar, Aniruddh Kurtkoti.An Approach To Graphics Passthrough In Cloud Virtual Machines[J]. International Journal of Engineering Research & Technology (S2278-0181), 2013, 2(6): 746-749.
[14] Shinpei Kato, Scott Brandt, Yutaka Ishikawa, et al. Operating Systems Challenges for GPU Resource Management[C]// 7th Annual Workshop on Operating Systems Platforms for Embedded Real-Time Applications. Porto, USA: George Washington University, 2011: 23-32.
[15] 齐平, 李龙澍. 云环境下结合模糊商空间理论的资源调度算法[J]. 小型微型计算机系统, 2013, 34(8): 1793-1797.
[16] 赵淦森, 虞海, 季统凯, 等. 云计算平台的自适应资源供给[J]. 电信科学, 2012, 28(1): 31-37.
[17] Y J Huang, T C Kuo, H K Lee.Fuzzy-PD Controller Design with Stability Equations for Electro-Hydraulic Servo Systems[C]// International Conference on Control, Seoul, Korea. USA: IEEE, 2007: 2407-2410.
[18] Dawei Sun, Guiran Chang, Changsheng Miao.Analyzing, Modeling and Evaluating Dynamic Adaptive Fault Tolerance Strategies in Cloud Computing Environments[J]. Journal of Supercomput (S0920-8542), 2013, 66(1): 193-228.
[19] Bo Yang, Feng Tan, Yuan-Shun Dai.Performance Evaluation of Cloud Service Considering Fault Recovery[J]. Journal of Supercomput (S0920-8542), 2013, 65(1): 426-444.
[20] 刘云生. 大规模分布式仿真系统容错关键技术研究[D]. 长沙: 国防科学技术大学, 2006.