vix.ing · top · new · best · stats · spec

Empowering In-Memory Relational Database Engines with Native Graph\n Processing

2017/09/19 by Mohamed S. Hassan, Hassan, Mohamed S., Tatiana Kuznetsova +7
Computer Science · #Advanced Database Systems and Queries #Data Management and Algorithms #Databases (cs.DB) #Distributed systems and fault tolerance #FOS: Computer and information sciences #Graph Theory and Algorithms

paper · pdf · doi:10.48550/arxiv.1709.06715

openalex publication_date 2017/09/19 · openalex created_date 2022/10/03 · openalex updated_date 2026/07/28

Abstract

The plethora of graphs and relational data give rise to many interesting\ngraph-relational queries in various domains, e.g., finding related proteins\nsatisfying relational predicates in a biological network. The maturity of\nRDBMSs motivated academia and industry to invest efforts in leveraging RDBMSs\nfor graph processing, where efficiency is proven for vital graph queries.\nHowever, none of these efforts process graphs natively inside the RDBMS, which\nis particularly challenging due to the impedance mismatch between the\nrelational and the graph models. In this paper, we propose to treat graphs as\nfirst-class citizens inside the relational engine so that operations on graphs\nare executed natively inside the RDBMS. We realize our approach inside VoltDB,\nan open-source in-memory relational database, and name this realization\nGRFusion. The SQL and the query engine of GRFusion are empowered to\ndeclaratively define graphs and execute cross-data-model query plans formed by\ngraph and relational operators, resulting in up to four orders-of-magnitude in\nquery-time speedup w.r.t. state-of-the-art approaches.\n

Related