Abstract: Facing the scenarios of distributed storage and cross-domain transfer of data in various business domains in industrial scenarios, this paper proposes a cloud-edge collaborative data storage and retrieval architecture to address the problems of multiple and complex business systems, huge amounts of data, and the inability of some data to be uploaded to the cloud, aiming to achieve unified storage of large-scale data and efficient cross-domain circulation. This architecture designs data encoding rules based on the RDF (Resource Description Framework) graph model and a multi-level efficient data storage strategy based on S-tree (Signature-tree) to ensure that data that cannot be uploaded to the cloud is stored in the edge server, and data that can be uploaded to the cloud is stored in the cloud server. It also proposes a cloud-edge oriented approach. Collaborative storage index tree efficient collaborative retrieval method (cloud-edge collaboration index) effectively improves the efficiency of data retrieval through the concurrent retrieval method and cloud-edge collaborative indexing method. Through experimental comparison with the RDF-3x and GRIN, the architecture proposed in this article can achieve higher performance in terms of operating efficiency and CPU utilization.

Key words: distributed storage, RDF &#40, Resource Description Framework&#41, graph model, cloud-edge collaboration

摘要： 面向工业场景各业务域中数据分布式存储和跨域流转使用的场景，针对业务系统多而复杂、数据量庞大且部分数据不可上云的问题，本文提出了云边协同的数据存储与检索架构，旨在实现大规模数据统一存储和跨域高效流转。该架构设计了基于RDF(resource description framework)图模型的数据编码规则和基于S-tree(signature-tree)的数据多层级高效存储策略，保证不可上云数据存储在边缘侧服务器，可上云数据存储在云服务器，并提出了面向云边协同存储的索引树高效协同检索方法(cloud-edge collaboration index)；通过并发检索机制和云边协同索引机制有效提高数据检索的效率；通过与RDF-3x、GRIN等方法进行实验对比，本文所提架构能够在运行效率和CPU利用率等方面都具有更高的效能。

关键词: 分布式存储, RDF模型, 云边协同, 协同存储, 协同检索