摘要
介绍了传统影像存储与传输系统(PACS)中数据存储所面临的挑战,分析了云存储在解决存储扩展、系统容错方面的天然优势,以及块存储的读写效率高、时延低等特点,以解决医疗影像数据长期存储的难题。通过采用混合云架构对影像数据进行分级存储,医院侧部署PACS应用服务器并通过磁盘阵列存储“热数据”,在云端使用基于IP存储网络(IP-SAN)的块存储网络来存储“冷数据”,定期将在线存储“热数据”迁移至云端“冷数据”资源池进行归档,节省了医院数据中心的存储空间。另外,通过块存储标签技术结合HASH校验技术,解决了院内“热数据”向云端“冷数据”迁移过程中确保数据一致性的难题。
The paper introduces the challenges faced by data storage in traditional image storage and transmission systems(PACS),analyzes the natural advantages of cloud storage in solving the problems of storage expansion and system fault tolerance,as well as the characteristics of block storage,such as high read/write efficiency and low latency,so as to solve the problem of long-term storage of medical image data.By adopting a hybrid cloud architecture for hierarchical storage of imaging data,PACS application servers are deployed at the hospital side,and"hot data"is stored through disk array,while"cold data"is stored through block storage network based on IP storage network(IP�SAN)in the cloud.Regular migration of online"hot data"to the cloud-based"cold data"resource pool is performed for archiving purposes,which can effectively save storage space in hospital data centers.In addition,based on the block storage labeling technology and HASH verification technology,the problem of how to ensure data consistency during the migration of"hot data"from the hospital to"cold data"in the cloud can be solved.
作者
姚恒斌
乔雅莉
YAO Hengbin;QIAO Yali(Information Department,Women&Children's Yangzhou Healthcare Hospital,Yangzhou 225002,Jiangsu Province,China)
出处
《中国数字医学》
2024年第2期99-103,共5页
China Digital Medicine
基金
江苏省高校哲学社会科学研究一般项目(2021SJA2438)。
关键词
影像云
块存储
归档
分级存储
Image cloud
Block storage
Archive
Hierarchical storage