[Paper Review] Simplicial Databases

David I. Spivak|ArXiv.org|Apr 13, 2009

Data Management and Algorithms7 references11 citations

TL;DR

This paper introduces simplicial databases—a categorical framework where database schemas are modeled as simplicial sets, endowing them with geometric structure to better represent relationships between tables. By treating data as sheaves over these schemas and defining morphisms that preserve structure, the approach enables limits and colimits to correspond naturally to standard database operations like joins and unions, offering a more geometric and flexible foundation than the traditional relational model.

ABSTRACT

In this paper, we define a category DB, called the category of simplicial databases, whose objects are databases and whose morphisms are data-preserving maps. Along the way we give a precise formulation of the category of relational databases, and prove that it is a full subcategory of DB. We also prove that limits and colimits always exist in DB and that they correspond to queries such as select, join, union, etc. One feature of our construction is that the schema of a simplicial database has a natural geometric structure: an underlying simplicial set. The geometry of a schema is a way of keeping track of relationships between distinct tables, and can be thought of as a system of foreign keys. The shape of a schema is generally intuitive (e.g. the schema for round-trip flights is a circle consisting of an edge from $A$ to $B$ and an edge from $B$ to $A$), and as such, may be useful for analyzing data. We give several applications of our approach, as well as possible advantages it has over the relational model. We also indicate some directions for further research.

Motivation & Objective

• To develop a mathematically rigorous, category-theoretic foundation for databases that improves upon the limitations of the classical relational model.
• To endow database schemas with geometric structure using simplicial sets, enabling intuitive visualization of table relationships.
• To define morphisms between databases that preserve both schema structure and data, facilitating reasoning about schema evolution and integration.
• To show that standard database operations (e.g., join, union) arise naturally as categorical constructions (limits and colimits) in the proposed framework.
• To lay the groundwork for future extensions involving functional dependencies, normalization, type changes, and ontology integration.

Proposed method

• Model databases as sheaves of sets over simplicial sets that represent schemas, using the language of category theory and algebraic topology.
• Define the category of tables as a foundational layer, with limits and colimits corresponding to natural join and union operations.
• Construct the category of simplicial databases as a full subcategory of a larger category of structured data, ensuring closure under all limits and colimits.
• Use the Yoneda lemma and categorical duality to formalize data preservation and morphism behavior across schemas.
• Introduce internal keys as a separate sheaf to distinguish between intrinsic data and management identifiers, improving schema integration.
• Generalize type specifications from functions between sets to functors between categories, enabling richer data modeling including orderings, constraints, and causal relationships.

Experimental results

Research questions

• RQ1 Can database schemas be given a natural geometric structure that reflects table relationships and improves query design?
• RQ2 Do categorical constructions like limits and colimits in the category of simplicial databases correspond precisely to standard database operations such as select, join, and union?
• RQ3 Can the geometric nature of simplicial schemas help visualize and reason about functional dependencies and normalization?
• RQ4 How can morphisms in the category of simplicial databases support schema and data integration while distinguishing between intrinsic data and arbitrary identifiers?
• RQ5 Can the framework be extended to model constraints, type changes, and causal relationships through enriched type specifications?

Key findings

• The category of simplicial databases is closed under all limits and colimits, and these constructions correspond exactly to standard database operations such as natural join, union, and selection.
• Schemas in the simplicial database model are naturally equipped with geometric structure via simplicial sets, allowing intuitive visualizations of relationships (e.g., round-trip flights as a 1-cycle).
• Data is modeled as a sheaf over the schema, enabling functorial transport of data between schemas and supporting formal reasoning about data consistency and transformation.
• Morphisms in the category preserve both schema structure and data, providing a rigorous foundation for database integration and schema evolution.
• The use of internal keys as a separate sheaf allows for clear separation between intrinsic data and management identifiers, improving clarity in integration and normalization tasks.
• Generalizing type specifications to functors between categories allows modeling of complex data types, orderings, constraints, and causal relationships, enhancing expressiveness beyond the relational model.