一种面向边缘计算的可扩展网络I/O框架

作为当前的研究热点,边缘计算可以在网络边缘部署计算、存储和网络资源,为用户提供云化服务,以满足新兴应用的处理需求。但是,当前网络I/O模型的扩展性问题使得运行在边缘节点上的边缘应用难以有效地利用丰富的边缘物理资源处理海量用户的网络请求。针对边缘计算场景下网络I/O模型的数据竞争和负载不均衡问题,提出了一种新的可扩展网络I/O模型。通过试验可以看出,利用该模型的边缘应用具有良好的性能和可扩展性。

A scalable network I/O framework for edge computing

As a hot research topic,edge computing deploys computing,storage and network resources at the edge of network to provide cloud services for users,and satisfies the requirement of emerging applications.However,the edge applications running on the edge server are diffcult to utilize physical resources to handle the network requests of edge users due to the scalability problem of current network I/O models.This paper proposes a new scalable network I/O model for edge applications to resolve the data race and load imbalance problems of current network I/O models.Evaluation shows that the edge application using our model achieves better performance and scalability.