面向工业物联网的分布式数据存储分配方案  

Distributed Data Storage Allocation Scheme for the Industrial Internet of Things

作　　者：丛犁 李锐 王华 任智源[3] CONG Li;LI Rui;WANG Hua;REN Zhiyuan(Information Communication Company,State Grid Jilin Electric Power Co.,Ltd.,Changchun 130000,China;State Grid Information Communication Group Co.,Ltd.,Beijing 100070,China;Xidian University,Xi’an 710071,China)

机构地区：[1]国网吉林省电力有限公司信息通信公司,吉林长春130000 [2]国网信息通信产业集团有限公司,北京100070 [3]西安电子科技大学,陕西西安710071

出　　处：《电力信息与通信技术》2020年第7期52-57,共6页Electric Power Information and Communication Technology

基　　金：科技部重点研发计划(2016YFE0123000)。

摘　　要：工业物联网(Industrial Internet of Things,IIoT)是实施智能制造的重要载体,近年来引起了学术界和业界的极大兴趣。目前,在IIoT环境中主要采用云存储来存储数据,但是,云存储的过度延迟无法满足IIoT业务的时延要求,尤其是某些时延敏感型的业务。针对云存储应用于IIoT大数据场景存在数据传输时延较高的问题,文章研究了一种新型的"云/雾混合"网络架构,该架构利用雾计算层设备构建分布式雾存储体系结构,通过将IIoT中的数据分布式存储至多个雾节点,达到降低数据传输时延的目的。由于雾设备的可靠性低,数据容易丢失,进一步提出了分布式冗余存储分配方案,以提高存储系统的安全性。由于提出的问题是所有非确定性多项式(Non-deterministic Polynomial,NP)问题,是能在多项式时间复杂度内归约到的问题,因此提出一种启发式优化算法,用于在安全约束下最小化数据传输时延。性能评估结果验证了所提方案能够降低数据传输时延,同时保证雾存储系统的安全性。Industrial Internet of Things (IIoT) is an important carrier for implementing intelligent manufacturing and has drawn great interests of both the academia and industry in recent years.Currently,the main-stream IIoT storage approach is almost cloud storage.However,the excessive delay of cloud storage cannot satisfy the latency requirements of IIoT business and even result in the failure of the manu-facturing,process,especially for some latency-sensitive business.Thus,a novel hybrid cloud and fog network architecture is proposed to solve the problem of high transmission latency of IIoT big data in cloud storage.By distributing the data on multiple fog devices,which are closer to the intelligent manufacturing equipment,the transmission latency can be greatly reduced.However,due to the weak reliability of fog devices,data is easily lost.Thus,we proposed a distributed redundant storage allo-cation scheme to enhance the security of the storage system.Since the optimization problem formu-lated is NP-hard(non-deterministic polynomial),a meta-heuristic algorithm is conceived for mini-mizing the latency with security constraints.Our performance evaluation results validate that the proposed scheme is indeed capable of reducing the latency as well as guaranteeing the security of the fog storage system.

关 键 词：工业物联网 雾计算 云/雾混合架构 数据安全存储 数据冗余 数据分配 

分 类 号：TP39[自动化与计算机技术—计算机应用技术]