PaaS架构后端管理平台的云边协同调度算法设计

Design of cloud side collaborative scheduling algorithm for PaaS architecture back⁃end management platform

PaaS架构后端管理平台容器资源由多模块选取,调度任务协同度不高,很难形成统一的调度方案。为此,文中设计一种基于云边协同的PaaS架构后端管理平台调度算法。通过分析PaaS架构后端管理平台资源的特殊性,构建代码构建模块、服务部署模块、后端服务模块的统一云数据特征;再基于相关函数完成模块统一云边数据的特征提取。通过云边协同计算框架进行资源部署和任务调度,优先将云服务中心的预测结果推送至最佳服务器边缘,提升云平台响应及资源利用率。利用帕累托最优对任务进行预测,实现PaaS架构后端管理平台任务的统一调度。实验结果表明,所提算法能够提升PaaS架构后端管理平台的资源利用率及平台响应时效性,满足用户需求。

The container resources of the back-end management platform of PaaS architecture are selected by multiple modules,and the coordination degree of scheduling tasks is not high,making it difficult to form a unified scheduling solution.A cloud-side collaboration based scheduling algorithm for PaaS architecture back-end management platform is proposed.The particularity of PaaS architecture back-end management platform resources is analyzed,and the unified cloud data characteristics of code construction module,service deployment module and back-end service module are constructed.The feature extraction of unified cloud edge data is completed based on correlation function.Resource deployment and task scheduling are carried out by means of the cloud edge collaborative computing framework,and the prediction results in the cloud service center are pushed to the best server edge first,so as to improve the response and resource utilization of the cloud platform.The Pareto algorithm is used to predict tasks,and the unified task scheduling of the PaaS architecture back-end management platform is realized.The experimental results show that the proposed algorithm can improve the resource utilization of PaaS back-end management platform and the timeliness of platform response,and meet user needs.