Heterogenous Data Source Exploring and Tracking in the PolyDBMS architecture (Bachelor Thesis, Ongoing)
Author
Description
The data management landscape has diversified significantly since its inception. Today, a vast array of database systems exists, each offering distinct capabilities inspired by varying data models and query languages. To address this heterogeneity, traditional databases have evolved to extend their original model definitions.
Parallel to this evolution, Multi-Model Databases have gained significant traction. Unlike legacy systems that retroactively add support for new models, these systems are designed to support multiple data models from the outset. While various architectures exist to achieve this, most rely on a single engine capable of handling different data models in parallel. However, managing multiple models simultaneously is complex; these systems often rely on a "primary model" and language, extending it to accommodate others, which can limit the native strengths of the secondary models.
An alternative approach is the multi-model database relying on multiple, separate single-model engines. One such architecture is the PolyDBMS architecture and its implementation Polypheny, which clearly separates different data models and query languages. However, this architecture requires significant effort to manage the collection of heterogeneous underlying engines.
The objective of this thesis is to conceptualize and implement a generic approach for the dynamic exploration, schema tracking, and synchronization of externally managed data engines (Sources) within the Polypheny ecosystem. The focus lies on Source, which follow one of the three supported data models natively.
Start / End Dates
2026/03/16 - 2026/07/15