Graph data processing method and device, electronic equipment and storage medium

By adopting a semi-external storage model in graph data processing, the storage node data is in memory and edge data is in persistent storage space, the problem of high memory and computing cost in large-scale graph data processing is solved, and efficient super-large-scale graph segmentation is achieved.

CN119719420APending Publication Date: 2025-03-28GUANGDONG OPPO MOBILE TELECOMMUNICATIONS CORP LTD +1
0 Cites 0 Cited by

Patent Information

Application Number
CN202311278652.7
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2023-09-28
Publication Date
2025-03-28

AI Technical Summary

Technical Problem

In large-scale graph data processing scenarios, the Metis algorithm requires high memory and high computing costs, resulting in low graph data processing efficiency.

Method used

By storing the node data of each node in the graph data in memory, the edge data of each edge connected to each node is stored in a persistent storage space, and using a semi-external storage model, the memory required for the operation of the graph segmentation system is reduced, and by reasonably arranging the data access method of edge data, the ultra-large-scale graph segmentation is achieved under memory limitations.

Benefits of technology

This greatly reduces the memory required for the operation of the graph segmentation system, and completes super-large-scale graph segmentation with reasonable time under memory limits, improving the efficiency of graph data processing.

✦ Generated by Eureka AI based on patent content.
Patent Text Reader

Abstract

The invention provides a graph data processing method and device, electronic equipment and a storage medium, and belongs to the technical field of computers. The method comprises the following steps: acquiring initial graph data, wherein the initial graph data comprises node data of V nodes and edge data of E edges; for any node in the V nodes, node data of the node is stored in a memory, target edge data associated with the node is stored in a persistent storage space according to the node number of the node, and the target edge data is edge data of at least one edge connected with the node in the E edges; and based on the node data in the memory and the edge data read from the persistent storage space, segmenting the initial graph data until a segmentation result of the initial graph data reaches a segmentation target. According to the scheme, the memory required by the operation of the graph segmentation system can be greatly reduced, and the super-large-scale graph segmentation can be completed within reasonable time under the memory limitation.
Need to check novelty before this filing date? Find Prior Art