面向医疗大数据的云雾网络及其分布式计算方案

A Cloud and Fog Network Architecture for Medical Big Data and Its Distributed Computing Scheme

针对云计算应用于医疗大数据场景时存在业务处理时延较高的问题,提出了一种基于边缘计算的新型云/雾混合网络架构,该架构利用医院中的路由器或交换机等边缘设备,在云服务器与医疗检测设备之间构建一个雾计算层,通过将云服务器中的医学影像等医疗大数据分析结果主动缓存至雾计算设备,并与雾设备上来自医疗检测终端的数据进行对比计算,得出诊断结果,达到降低业务处理时延的目的。考虑到边缘设备的计算能力较弱,进一步提出了一种多设备分布式计算方案,利用带约束的粒子群优化负载均衡（CPSO-LB）算法,达到任务处理时延最小的目标。仿真结果表明：基于CPSO-LB算法的云/雾混合网络能有效地降低医疗数据处理时延;当采用10个雾计算设备,处理的医疗数据量在6-10Gb时,与云计算网络相比时延性能提升了50.95%-37.37%。

A novel hybrid cloud and fog network architecture based on edge computing is proposed to solve the problem of high processing latency of the medical big data in cloud computing network. The architecture adds a fog computing layer between cloud servers and medical measurement devices by using edge devices such as routers or switches in a hospital. Fog computing devices proactively cache analysis results of medical images and other medical big data from cloud servers, and compare these data with the data from medical measurement devices to get the diagnostic results and reduce processing latency. Meanwhile, a multi device distributed computing scheme is proposed by considering the weak computing power of edge devices and a constrained particle swarm optimization load balancing （CPSO-LB） algorithm is applied to minimize the latency. Simulation results indicate that the novel network architecture with CPSOLB algorithm decreases the latency effectively. A comparison with a cloud computing shows that it＇s latency performance increases by 50.95%-37.37% when 10 fog devices and processing 6-10 Gb medical data are used.