Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

A Multi-GPU Based Strongly Connected Graph Detection Method

A technology of strongly connected graphs and detection methods, applied in image memory management, processor architecture/configuration, etc., can solve problems such as limiting efficiency and scope of application, affecting algorithm execution efficiency, unable to fully utilize GPU, etc., to achieve the effect of optimized design

Active Publication Date: 2021-05-04
INST OF INFORMATION ENG CHINESE ACAD OF SCI
View PDF3 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, GPU-based strongly connected graph detection schemes all use a single GPU for processing, and cannot make full use of all GPUs on the device, or use existing graph computing systems for strongly connected graph detection, which will bring a large amount of data interaction between GPUs , which ultimately affects the execution efficiency of the algorithm
Therefore, parallel schemes for strongly connected graph detection using a single GPU or using a graph system limit the efficiency and applicability of detection on large graphs

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • A Multi-GPU Based Strongly Connected Graph Detection Method
  • A Multi-GPU Based Strongly Connected Graph Detection Method
  • A Multi-GPU Based Strongly Connected Graph Detection Method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0077] In order to make the above features and advantages of the present invention more comprehensible, the present invention will be further described in detail below through specific embodiments and accompanying drawings.

[0078] This embodiment proposes a method for detecting strongly connected graphs based on multiple GPUs. This method fully utilizes the parallel computing capabilities of multiple GPUs, and incorporates multiple algorithms, and uses different algorithms according to task requirements at different stages. Such as figure 1 As shown, the implementation process of the present invention is divided into two parts: running processing on the CPU and running processing on multiple GPUs. For a more detailed division, the entire implementation process can be decomposed into nine steps. The implementation of each step is described in detail below:

[0079] Step 1: Load graph data and convert it into a unified storage format:

[0080] Read the graph data information...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The present invention proposes a multi-GPU based strongly connected graph detection method, comprising the following steps: loading graph data and unifying the storage format; preprocessing the graph data, including partitioning the graph according to the number of partitions and performing partition storage, and interlinking The vertices in different partitions are copied for vertex processing; the preprocessed data is stored in multiple GPUs, and the breadth-first traversal is performed centering on the copied vertices and the copied edge information is recorded; the copied edges are sent back to the CPU to detect the strongly connected graph And mark the vertices belonging to the same strongly connected graph; send the marked vertices back to the above-mentioned multiple GPUs to perform strongly connected graph detection.

Description

technical field [0001] The invention relates to a method for detecting strongly connected graphs using a heterogeneous system, in particular to a method for detecting strongly connected graphs based on multiple GPUs. Background technique [0002] As a basic data structure, the graph data structure has been widely used because it can well express the correlation between data. With the development of artificial intelligence and social networks, accelerated computing of large amounts of graph data has become a hot field. The strongly connected graph (Strongly Connected Components, SCC) is a basic graph structure. How to quickly detect the strongly connected graph of graph data is the basic problem of many graph computing applications. fields have important applications. [0003] Strongly connected graph detection is a research direction that started very early. Before general-purpose GPUs were widely used, researchers generally used CPUs for strongly connected graph detectio...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Patents(China)
IPC IPC(8): G06T1/20G06T1/60
CPCG06T1/20G06T1/60
Inventor 吴广君王树鹏侯骏腾李斌斌
Owner INST OF INFORMATION ENG CHINESE ACAD OF SCI
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com