Textual and Visual Algebra Exploration (Master Project, Ongoing)


Tobias Martin Weber


The relational algebra, introduced by Edgar F. Codd in 1970, stands as a fundamental
pillar of the theory of relational database systems. It provides a set of operations for
manipulating and querying the data within a relational database system. Multiple of these
operations are combined into a tree structure, also referred to as a query plan.

In Polypheny, a heavily extended version of this relational algebra is being used. Besides
the relational operators, the algebra used by Polypheny also includes document and graph
operators. This extended algebra is called PolyAlgebra and is used by Polypheny as basis
for its entire query processing, optimization, and execution.

Due to its architecture, Polypheny distinguishes between three sub-algebras of this PolyAlgebra.
Following the same general structure, the LogicalAlgebra is used to represent the
query with regard to the logical schema. The AllocationAlgebra is used in the query routing
phase. The PhysicalAlgebra represents the query based on the data model, capabilities
and available set of operators on the underlying data stores. Thus, for every operator in
the LogicalAlgebra, there might be multiple counterparts in the PhysicalAlgebra.

Currently, Polypheny features an algebra builder named AlgBuilder and two separate
systems for visualizing and editing query plans. However, these systems only cover a
limited subset of the operations. Furthermore, they are incompatible with each other.
The primary objective of this thesis is to expose the algebra directly to the user, allowing
them to view and edit query plans expressed by the PolyAlgebra. This interaction shall
be facilitated not only through a well-defined textual representation but also through
visual means, providing a comprehensive understanding of the underlying structures and

Start / End Dates

2024/03/04 - 2024/07/17