A workload-driven approach for automatic schema generation for document stores
Neha Bansal, Shelly Sachdeva, Lalit Kumar Awasthi · 2023
NoSQL data modeling issues always rely on rules of thumb and require expert assistance to design the schema in a manner that optimizes the performance of the application. This paper seeks to develop a schema design model for a NoSQL database with the vision of automating the schema design process that is very time-consuming with the manual rule-based approaches. The proposed model takes input from the conceptual model (EER model) and application workload in the form of queries and anticipated data volume. The characteristics of application workload are studied in the form of labels. The conceptual model is converted into a physical schema design for document stores with the aid of generated labels. The proposed work reduced the overall time, cost and efforts needed to create or correct subpar schema.