Zobrazit v EDS

Accelerating complex graph queries by summary-based hybrid partitioning for discovering vulnerabilities of distribution equipment.

Uloženo v:
Podrobná bibliografie
Název: Accelerating complex graph queries by summary-based hybrid partitioning for discovering vulnerabilities of distribution equipment.
Autoři: Wang, Qiong1,2 (AUTHOR), He, Wei1 (AUTHOR), Yang, Shang3 (AUTHOR), Zhao, Ruoyu1 (AUTHOR), Ma, Yinglong1 (AUTHOR) yinglongma@ncepu.edu.cn
Zdroj: Future Generation Computer Systems. Jun2025, Vol. 167, pN.PAG-N.PAG. 1p.
Témata: *ELECTRIC power distribution grids, *SQL, *INFORMATION sharing
Abstrakt: With the high proportion of electrical and electronic devices in China's power grids, massive graph data of power distribution equipment has been accumulated to share the knowledge across heterogeneous information, while the vulnerabilities of power devices consequently trigger new security risks to the power grid. It is crucial to swiftly and accurately discover the intrinsic vulnerabilities of power devices from the massive power distribution graph data for ensuring safe operation of the power grid. However, diverse complex queries make it inefficient to achieve consistent graph querying performance over the massive power graph data for swift and accurate vulnerability discovery in a highly available and user-friendly manner. To handle the aforementioned problem, in this paper, we present a power graph query-oriented pipeline framework to consistently accelerate complex graph queries over the massive graph data of power distribution equipment for efficient vulnerability discovery. First, we propose a lossless graph summarization method, through which a summary graph is produced from the raw graph data. Second, very different from existing methods, we propose a two-stage hybrid partitioning including the binary partitioning and the consequent ternary partitioning, which is conducted based on the summary graph instead of the raw graph for reducing the search scope and minimizing the input of the queried data, thereby accelerating the query. Third, the complex graph query with multiple triplet patterns will be automatically translated into the Spark SQL statement for query execution without users' interference, through which the accurate results will be obtained by recovering the summary-based intermediate results. At last, extensive experiments were made over four datasets against some state-of-the-art methods, and the results show that our approach is very competitive with these approaches and achieves consistent graph querying performance in accelerating complex graph queries while obtaining accurate results. • A power graph query-oriented pipeline framework for vulnerability discovery. • A lossless summarization method to produce the summary graph. • A two-stage summary-based hybrid partitioning approach. • Extensive experiments were made over four datasets to validate the efficiency. [ABSTRACT FROM AUTHOR]
Copyright of Future Generation Computer Systems is the property of Elsevier B.V. and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)
Databáze: Business Source Index
Popis
Abstrakt:With the high proportion of electrical and electronic devices in China's power grids, massive graph data of power distribution equipment has been accumulated to share the knowledge across heterogeneous information, while the vulnerabilities of power devices consequently trigger new security risks to the power grid. It is crucial to swiftly and accurately discover the intrinsic vulnerabilities of power devices from the massive power distribution graph data for ensuring safe operation of the power grid. However, diverse complex queries make it inefficient to achieve consistent graph querying performance over the massive power graph data for swift and accurate vulnerability discovery in a highly available and user-friendly manner. To handle the aforementioned problem, in this paper, we present a power graph query-oriented pipeline framework to consistently accelerate complex graph queries over the massive graph data of power distribution equipment for efficient vulnerability discovery. First, we propose a lossless graph summarization method, through which a summary graph is produced from the raw graph data. Second, very different from existing methods, we propose a two-stage hybrid partitioning including the binary partitioning and the consequent ternary partitioning, which is conducted based on the summary graph instead of the raw graph for reducing the search scope and minimizing the input of the queried data, thereby accelerating the query. Third, the complex graph query with multiple triplet patterns will be automatically translated into the Spark SQL statement for query execution without users' interference, through which the accurate results will be obtained by recovering the summary-based intermediate results. At last, extensive experiments were made over four datasets against some state-of-the-art methods, and the results show that our approach is very competitive with these approaches and achieves consistent graph querying performance in accelerating complex graph queries while obtaining accurate results. • A power graph query-oriented pipeline framework for vulnerability discovery. • A lossless summarization method to produce the summary graph. • A two-stage summary-based hybrid partitioning approach. • Extensive experiments were made over four datasets to validate the efficiency. [ABSTRACT FROM AUTHOR]
ISSN:0167739X
DOI:10.1016/j.future.2025.107747