Please wait a minute...
Front. Inform. Technol. Electron. Eng.  2014, Vol. 15 Issue (1): 13-30    DOI: 10.1631/jzus.C1300121
    
Activity-based simulation using DEVS: increasing performance by an activity model in parallel DEVS simulation
Bin Chen, Lao-bing Zhang, Xiao-cheng Liu, Hans Vangheluwe
College of Information System and Management, National University of Defense Technology, Changsha 410073, China; Department of Mathematics and Computer Science, University of Antwerp, Antwerp 2020, Belgium; School of Computer Science, McGill University, Montréal H3A 2A7, Canada
Download:   PDF(0KB)
Export: BibTeX | EndNote (RIS)      

Abstract  Improving simulation performance using activity tracking has attracted attention in the modeling field in recent years. The reference to activity has been successfully used to predict and promote the simulation performance. Tracking activity, however, uses only the inherent performance information contained in the models. To extend activity prediction in modeling, we propose the activity enhanced modeling with an activity meta-model at the meta-level. The meta-model provides a set of interfaces to model activity in a specific domain. The activity model transformation in subsequence is devised to deal with the simulation difference due to the heterogeneous activity model. Finally, the resource-aware simulation framework is implemented to integrate the activity models in activity-based simulation. The case study shows the improvement brought on by activity-based simulation using discrete event system specification (DEVS).

Key wordsActivity tracking      Activity enhanced modeling      Discrete event system specification (DEVS)      Resource-aware simulation framework     
Received: 09 May 2013      Published: 07 January 2014
CLC:  TP312  
Cite this article:

Bin Chen, Lao-bing Zhang, Xiao-cheng Liu, Hans Vangheluwe. Activity-based simulation using DEVS: increasing performance by an activity model in parallel DEVS simulation. Front. Inform. Technol. Electron. Eng., 2014, 15(1): 13-30.

URL:

http://www.zjujournals.com/xueshu/fitee/10.1631/jzus.C1300121     OR     http://www.zjujournals.com/xueshu/fitee/Y2014/V15/I1/13


基于活跃度的DEVS仿真优化方法:在并行DEVS仿真中依靠活跃度模型提高性能

研究目的:研究如何用模型活跃度提高仿真性能,包括对模型的活跃度进行量化,将多范式建模方法用于构造领域活跃度模型等,以及构造支持活跃度模型的建模与仿真框架问题。
创新方法:已有的仿真性能优化方法专注于对性能运行结果的“盲”挖掘,而忽略了模型本身可以提供的性能信息。本文利用活跃度的概念在建模阶段就提取模型的性能信息,用于在仿真阶段指导计算资源的分配。此外,利用活跃度元模型和模型变换方法,研究在任意领域进行活跃度建模的方法,基于DEVS仿真引擎提出仿真中活跃度模型异质的解决方法。
研究手段:利用CPU占用率和内存使用量来量化模型的活跃度,解决了活跃度模型的计算问题。设计了增强式活跃度建模方法,提供了一系列特定接口对特定领域上的模型活跃度进行构造。提出了资源敏感型的建模与仿真框架,为活跃度建模定义了一整规范,选择DEVS作为通用仿真形式化方法,方便用户构造各种类型的领域活跃度模型。
重要结论:对如何利用活跃度进行仿真性能优化进行了研究,利用计算机资源占用的方法量化了模型的活跃度,提供了一种在特定领域中定义活跃度的增强式活跃度建模方法,利用活跃度元模型和模型变换方法解决了仿真异质问题,设计了资源敏感型建模与仿真框架为在仿真应用中构造和使用活跃度模型提供了标准规范。

关键词: DEVS,  活跃度追溯,  活跃度增强式建模,  资源敏感 
[1] Xin Li, Jin Sun, Fu Xiao. An efficient prediction framework for multi-parametric yield analysis under parameter variations[J]. Front. Inform. Technol. Electron. Eng., 2016, 17(12): 1344-1359.
[2] Xin Li, Jin Sun, Fu Xiao, Jiang-shan Tian. An efficient bi-objective optimization framework for statistical chip-level yield analysis under parameter variations[J]. Front. Inform. Technol. Electron. Eng., 2016, 17(2): 160-172.
[3] Mei Wen, Da-fei Huang, Chang-qing Xun, Dong Chen. Improving performance portability for GPU-specific OpenCL kernels on multi-core/many-core CPUs by analysis-based transformations[J]. Front. Inform. Technol. Electron. Eng., 2015, 16(11): 899-916.
[4] Ye-tian Fan, Wei Wu, Wen-yu Yang, Qin-wei Fan, Jian Wang. A pruning algorithm with L1/2 regularizer for extreme learning machine[J]. Front. Inform. Technol. Electron. Eng., 2014, 15(2): 119-125.