GraphflowDB: A DBMS for large scale graph data management
GraphflowDB is an in-memory DBMS targeting primarily read-heavy subgraph query workloads equivalent to select-project-join (SPJ) queries over graph data. The system supports the openCypher language. Our research focuses on rethinking each core database component including query processing and optimization, storage layout and compression, and materialized views.
Talks
Pinterest Labs Tech Talk: "GraphflowDB and Modern Query Processing Techniques for Graph Database Systems" (March 2021)
Publications
- Accurate Summary-based Cardinality Estimation Through the Lens of Cardinality Estimation Graphs (PVLDB 2022)
- Columnar Storage and List-based Processing for Graph Database Management Systems (PVLDB 2021)
- Optimizing One-time and Continuous Subgraph Queries using Worst-Case Optimal Joins (TODS 2021)
- A+ Indexes: Tunable and Space-Efficient Adjacency Lists in Graph Database Management Systems (ICDE 2021)
- R2GSync and Edge Views: Practical RDBMS to GDBMS Synchronization (GRADES-NDA 2021)
- Optimizing Subgraph Queries by Combining Binary and Worst-Case Optimal Joins (PVLDB 2019)
- The Ubiquity of Large Graphs and Surprising Challenges of Graph Processings: Extended Survey (VLDBJ 2019)
- The Ubiquity of Large Graphs and Surprising Challenges of Graph Processings (PVLDB 2018)
- Graphflow: An Active Graph Database (Demo @ SIGMOD 2017)