R2GSync and edge views: practical RDBMS to GDBMS synchronization

Graph databases that are used in enterprises are primarily extracted from a main transactional store that is often an RDBMS. This data infrastructure set up raises the challenge of keeping the extracted graph in a graph database management system (GDBMS) in sync with the source RDBMS. When the extracted graphs contain edge types that are results of join queries, this synchronization requires incrementally maintaining these join queries. In this paper, we investigate an alternative design where we can map the individual relations in these joins to virtual nodes and edges to keep the synchronization very efficient and instead support view-based querying in the GDBMS. We present a system called R2GSync, that synchronizes an RDBMS with a GDBMS and our accompanying edge view design for a GDBMS. We describe our implementation of edge views in GraphflowDB and query optimization techniques for improving the performance of queries that involve edge views.

[1]  Bruce G. Lindsay,et al.  How to roll a join: asynchronous incremental view maintenance , 2000, SIGMOD '00.

[2]  Sangkeun Lee,et al.  Table2Graph: A Scalable Graph Construction from Relational Tables Using Map-Reduce , 2015, 2015 IEEE First International Conference on Big Data Computing Service and Applications.

[3]  Amine Mhedhbi,et al.  The ubiquity of large graphs and surprising challenges of graph processing: extended survey , 2017, The VLDB Journal.

[4]  Walid G. Aref,et al.  GRFusion: Graphs as First-Class Citizens in Main-Memory Relational Database Systems , 2018, SIGMOD Conference.

[5]  Amol Deshpande,et al.  Extracting and Analyzing Hidden Graphs from Relational Databases , 2017, SIGMOD Conference.

[6]  Zhengping Qian,et al.  Real-time Constrained Cycle Detection in Large Dynamic Graphs , 2018, Proc. VLDB Endow..

[7]  Amine Mhedhbi,et al.  Integrating Column-Oriented Storage and Query Processing Techniques Into Graph Database Management Systems , 2021, ArXiv.

[8]  Yue Zhuge,et al.  Graph structured views and their incremental maintenance , 1998, Proceedings 14th International Conference on Data Engineering.

[9]  Jeremy Chen,et al.  Graphflow: An Active Graph Database , 2017, SIGMOD Conference.

[10]  Stefan Plantikow,et al.  Cypher: An Evolving Query Language for Property Graphs , 2018, SIGMOD Conference.

[11]  Frank Wm. Tompa,et al.  Efficiently updating materialized views , 1986, SIGMOD '86.

[12]  Gábor Szárnyas,et al.  Incremental View Maintenance for Property Graph Queries , 2017, SIGMOD Conference.

[13]  Semih Salihoglu,et al.  GraphWrangler: An Interactive Graph View on Relational Data , 2019, SIGMOD Conference.

[14]  Abdelkamel Tari,et al.  Materialized View Maintenance: Issues, Classification, and Open Challenges , 2019, Int. J. Cooperative Inf. Syst..

[15]  Jeffrey F. Naughton,et al.  A comparison of three methods for join view maintenance in parallel RDBMS , 2003, Proceedings 19th International Conference on Data Engineering (Cat. No.03CH37405).

[16]  Theodore L. Willke,et al.  GraphBuilder: scalable graph ETL framework , 2013, GRADES.

[17]  Jimmy J. Lin,et al.  Real-Time Twitter Recommendation: Online Motif Detection in Large Dynamic Graphs , 2014, Proc. VLDB Endow..

[18]  Carlo Curino,et al.  Kaskade: Graph Views for Efficient Graph Analytics , 2020, 2020 IEEE 36th International Conference on Data Engineering (ICDE).

[19]  Amine Mhedhbi,et al.  Optimizing Subgraph Queries by Combining Binary and Worst-Case Optimal Joins , 2019, Proc. VLDB Endow..

[20]  Udayan Khurana,et al.  GraphGen: Exploring Interesting Graphs in Relational Data , 2015, Proc. VLDB Endow..

[21]  Walid G. Aref,et al.  Extending In-Memory Relational Database Engines with Native Graph Support , 2018, EDBT.