Service-Oriented Model Encapsulation and Selection Method for Complex System Simulation Based on Cloud Architecture

• Main Authors: Siqi Xiong, Feng Zhu, Yiping Yao, Wenjie Tang, Yuhao Xiao
• Format: Article
• Language: English
• Published: MDPI AG 2019-09-01
• Series: Entropy
• Subjects: complex system simulation; cloud computing architecture; service-oriented modeling; semantic search framework; QoS-based service selection
• Online Access: https://www.mdpi.com/1099-4300/21/9/891

With the rise in cloud computing architecture, the development of service-oriented simulation models has gradually become a prominent topic in the field of complex system simulation. In order to support the distributed sharing of the simulation models with large computational requirements and to select the optimal service model to construct complex system simulation applications, this paper proposes a service-oriented model encapsulation and selection method. This method encapsulates models into shared simulation services, supports the distributed scheduling of model services in the network, and designs a semantic search framework which can support users in searching models according to model correlation. An optimization selection algorithm based on quality of service (QoS) is proposed to support users in customizing the weights of QoS indices and obtaining the ordered candidate model set by weighted comparison. The experimental results showed that the parallel operation of service models can effectively improve the execution efficiency of complex system simulation applications, and the performance was increased by 19.76% compared with that of scatter distribution strategy. The QoS weighted model selection method based on semantic search can support the effective search and selection of simulation models in the cloud environment according to the user’s preferences.