医疗健康物联网环境下云存储数据副本管理机制的设计与仿真实现
发布时间：2018-11-20 04:47
【摘要】：随着物联网技术的快速发展,各种基于物联网的应用已经深入到人们生活的各个领域,医疗健康便是其中的热点领域。医疗健康物联网环境下的数据具有海量、异构、难以管理等特点,而当前的数据副本管理机制结构简单、可扩展性差,不能为用户提供高效、可靠和安全的数据访问服务。数据副本管理机制是任何存储系统的重要组成部分,对医疗健康物联网环境下的云存储系统就更为重要。为了实现医疗数据存储的可行性、可靠性和容错性,本文设计了基于云存储的数据副本管理机制,包括数据副本的放置策略、选择策略和一致性维护策略。就数据副本的放置策略而言,根据医疗数据对访问时间不同层次的要求,提出了p-中心和p-中位模型；定义了新的邻域搜索函数,并设计了相应的组合算法；由于用户访问方式和网络状态的动态性,进而提出了数据副本调整策略。在考虑当前网络状态和存储节点负载的情况下,设计了基于模拟植物生长算法的数据副本选择策略,以提高用户对医疗数据的访问速度。考虑到用户对不同数据副本的读写操作会导致不一致问题,提出了数据副本的一致性维护策略,使每个数据副本都能响应访问请求,并通过时间戳机制避免数据副本的更新冲突问题,提高了系统的吞吐量和数据访问的可靠性。最后,对本文设计的基于云存储的数据副本管理机制进行仿真实现。性能评价结果表明,与原有的数据副本管理机制相比,本文设计的机制不仅可以降低数据访问的时间延迟和网络资源消耗,还实现了数据副本的负载均衡,同时提高了数据访问的可行性和可靠性。
[Abstract]:With the rapid development of the Internet of things technology, various applications based on the Internet of things have penetrated into all fields of people's lives, medical and health is one of the hot areas. The data in the medical and health Internet of things environment has the characteristics of magnanimity, heterogeneity and difficulty to manage. However, the current data replica management mechanism is simple in structure and poor in expansibility, which can not provide users with efficient, reliable and secure data access services. Data replica management mechanism is an important part of any storage system, especially for the cloud storage system in the medical and health Internet of things environment. In order to realize the feasibility, reliability and fault tolerance of medical data storage, this paper designs a data replica management mechanism based on cloud storage, including data replica placement strategy, selection strategy and consistency maintenance strategy. According to the requirements of medical data for different levels of access time, the p- center and p- median models are proposed, and a new neighborhood search function is defined, and a corresponding combination algorithm is designed. Because of the dynamic nature of user access mode and network state, a data replica adjustment strategy is proposed. Considering the current network state and storage node load, a data replica selection strategy based on simulated plant growth algorithm is designed to improve user access speed to medical data. Considering that users' reading and writing operations on different data replicas can lead to inconsistencies, a consistency maintenance strategy for data replicas is proposed to enable each data replica to respond to access requests. The update conflict of data replica is avoided by timestamp mechanism, and the system throughput and data access reliability are improved. Finally, the paper simulates the data replica management mechanism based on cloud storage. The performance evaluation results show that compared with the original data copy management mechanism, the mechanism designed in this paper can not only reduce the time delay of data access and network resource consumption, but also realize the load balancing of data copy. At the same time, the feasibility and reliability of data access are improved.
【学位授予单位】：东北大学
【学位级别】：硕士
【学位授予年份】：2013
【分类号】：TP333;TP391.44;TN929.5
本文编号：2343783
[Abstract]:With the rapid development of the Internet of things technology, various applications based on the Internet of things have penetrated into all fields of people's lives, medical and health is one of the hot areas. The data in the medical and health Internet of things environment has the characteristics of magnanimity, heterogeneity and difficulty to manage. However, the current data replica management mechanism is simple in structure and poor in expansibility, which can not provide users with efficient, reliable and secure data access services. Data replica management mechanism is an important part of any storage system, especially for the cloud storage system in the medical and health Internet of things environment. In order to realize the feasibility, reliability and fault tolerance of medical data storage, this paper designs a data replica management mechanism based on cloud storage, including data replica placement strategy, selection strategy and consistency maintenance strategy. According to the requirements of medical data for different levels of access time, the p- center and p- median models are proposed, and a new neighborhood search function is defined, and a corresponding combination algorithm is designed. Because of the dynamic nature of user access mode and network state, a data replica adjustment strategy is proposed. Considering the current network state and storage node load, a data replica selection strategy based on simulated plant growth algorithm is designed to improve user access speed to medical data. Considering that users' reading and writing operations on different data replicas can lead to inconsistencies, a consistency maintenance strategy for data replicas is proposed to enable each data replica to respond to access requests. The update conflict of data replica is avoided by timestamp mechanism, and the system throughput and data access reliability are improved. Finally, the paper simulates the data replica management mechanism based on cloud storage. The performance evaluation results show that compared with the original data copy management mechanism, the mechanism designed in this paper can not only reduce the time delay of data access and network resource consumption, but also realize the load balancing of data copy. At the same time, the feasibility and reliability of data access are improved.
【学位授予单位】：东北大学
【学位级别】：硕士
【学位授予年份】：2013
【分类号】：TP333;TP391.44;TN929.5
【参考文献】
相关期刊论文 前1条
1 刘田甜;李超;胡庆成;张桂刚;;云环境下多副本管理综述[J];计算机研究与发展;2011年S3期
相关博士学位论文 前1条
1 秦进;多商品物流网络设计相关优化模型及算法研究[D];中南大学;2006年
，本文编号：2343783
本文链接：https://www.wllwen.com/kejilunwen/jisuanjikexuelunwen/2343783.html
