When property graphs join SQL
summary
Oracle announces SQL/PGQ in Oracle 23c: property graphs become first-class SQL objects that map metadata over existing tables with no data copy, expose JSON fields as graph properties, and execute pattern and variable-length traversals by translating to SQL. Queries run via a new GRAPH_TABLE operator using PGQL-style patterns (example shows 1..3 hops to find friends-of-friends), integrate with APEX and materialized views, and leverage PGX and 60+ built-in algorithms for analytics while keeping transactional consistency.
AI-written summary · about
chapters (20)
- 0:00Introduction
- 0:23Connections are everywhere
- 0:52Graph is an important trend
- 1:32What are property graphs
- 2:25Property graphs vs relational models
- 3:26New use cases
- 4:39Multiple domains
- 4:54Why graph
- 6:51Oracle graph support
- 8:34What are SQL property graphs
- 11:07Create property graph syntax
- 12:22Explicit property graph syntax
- 13:47Graph table operator
- 14:53Graph query example
- 17:32Real life example
- 18:54Using graphs on top of relational data
- 21:57Querying with graphs
- 26:08Graphs are SQL objects
- 27:23How to build an Apex application
- 29:16Summary
related talks
27mExploring property graphs with SQL/PGQ in PostgreSQL
25mSimplify Property Graphs Using SQL with Database 23ai
45mGraph databases, PostgreSQL and SQL/PGQ
29mAll about graphs in Database
12mSolve Problems by Understanding Data and Graphs
06mUnlocking graph analytics in DuckDB with SQL/PGQ
29mA New Era for Graph Queries: ISO/IEC 39075 GQL
18mGraph analytics: From zero to hero