摘要： 随着云计算与空天地海一体化通信网络的深度融合,各种复杂应用场景的出现使得组合服务的种类和数量急剧增多,结构也变得复杂。在云网融合环境下,用户移动设备和边缘服务器等硬件能力有限,能耗问题成为组合服务进行动态重构不可忽略的重要因素。此外,传统方法并未考虑空天地海不同场景下用户对不同服务质量(QoS)属性需求的差异性,使得组合服务的交付结果并不令人满意。为了解决上述问题,提出一种基于多目标粒子群优化(PSO)的组合服务动态重构方法。该方法首先根据重构原子服务的三维空间地理位置和功能进行聚类,有效解决在云网融合环境下服务规模庞大情况下的搜索空间爆炸问题;然后通过能耗计算模型得到服务调用的综合能耗,并将其作为动态重构的优化目标之一,结合服务的多种QoS属性进行多目标寻优,最终生成符合用户需求且能耗较低的重构方案。实验结果表明,该方法在云网融合环境下节约能耗和应对较大候选服务集规模等方面具有较优性能。

关键词: 云网融合, 多目标粒子群优化算法, 组合服务, 动态重构, 服务质量

Abstract: The integration of cloud computing and communication networks in the space-air-ground-sea has led to complex application scenarios with a wide range of composite services. In cloud-network integration environments, the limited hardware capabilities of mobile user devices and edge servers make energy consumption a critical factor that cannot be ignored in the dynamic reconfiguration of composite services. Moreover, traditional methods have overlooked the varying needs of users in different scenarios space-air-ground-sea for different Quality of Service (QoS) attributes. This oversight has resulted in unsatisfactory delivery outcomes for composite services. To address this issue, this paper proposes a dynamic reconfiguration method using multi-objective Particle Swarm Optimization (PSO). This method clusters services based on the spatial location and function, effectively solving the problem of search space explosion in the context of large-scale service scale in cloud-network integration environments. The comprehensive energy consumption of service is obtained through an energy consumption calculation model and use it as one of the optimization objectives for dynamic reconstruction. Multiple particle fitness functions, considering various QoS attributes, are employed with an adaptive weight strategy for an efficient global search. Finally, the method generates a reconfiguration scheme that meets user requirements with low energy consumption. Experimental results demonstrate that the proposed method outperforms traditional approaches in terms of energy conservation and scalability for large candidate service scales in cloud-network integration environments.

Key words: cloud-network integration, multi-objective Particle Swarm Optimization (PSO) algorithm, composite service, dynamic reconfiguration, Quality of Service(QoS)

中图分类号: 

• TP311