Combining Two Types of Database System for Managing Property Graph Data

概要

A data structure called a graph consists of nodes and edges; a node represents an entity in data, and an edge represents a relationship between two entities. Moreover, graphs accepting the property graph model (PGM) can provide information such as a person's gender, age, home address. All graph database (GDB) generally support this graph model. Conventional GDBs cannot avoid scanning all the nodes or specified labeled ones from the database when the clients issue a query. This process is not efficient because queries scan unnecessary nodes when an executing a query that has conditions that it uses properties of graph components. In order to resolve this problem, we analyze how node and edge properties can be queried more efficiently while maintaining the structure of the graph. Therefore, in this study, we just propose an approach for graph data management, which separately stores nodes/edges information and property information into separate data structures. In addition, we describe a plan for conducting an evaluation of our approach because it is the proposal stage.

引用情報

Kazuma Kusu, Kenji Hatano, Combining Two Types of Database System for Managing Property Graph Data, Proceedings of 2018 IEEE International Conference on Big Data (Big Data), pp.5349-5351, 2018-12-10, DOI: 10.1109/BigData.2018.8622050.