Comparative Analysis of Relational and Graph Databases for Data Provenance: Performance, Queries, and Security Considerations

In today’s interconnected world, we perceive and understand everything as a vast network of relationships. Whether it’s people, places, or events, our minds naturally associate them as nodes and edges in a graph. This inherent cognitive tendency in human thinking contrasts with the formalisms commonly used by computer scientists and technologists, such as relational databases and JSON documents. While these abstractions serve as convenient tools, they only partially align with the intuitive way our brains process information. Relational databases excel in handling structured data and enforcing strict relationships, making them ideal for transaction processing and deterministic analytics. However, their rigid constraints often hinder the exploration of distant connections and answering complex questions about relationships that span multiple layers. This paper focuses on non-relational databases, particularly graph databases, and presents a comprehensive review of Neo4j compared to traditional relational databases. Graph databases offer a more realistic representation of corresponding data, allowing for flexible querying and traversing relationships with ease. By embracing the graph model, we can unlock new insights and uncover hidden patterns within our interconnected world.