TITLE:
Tailored Partitioning for Healthcare Big Data: A Novel Technique for Efficient Data Management and Hash Retrieval in RDBMS Relational Architectures
AUTHORS:
Ehsan Soltanmohammadi, Neset Hikmet, Dilek Akgun
KEYWORDS:
Healthcare Data Partitioning, Relational Database Management Systems (RDBMS), Big Data Management, Load Balance, Query Performance Improvement, Data Integrity and Fault Tolerance, Efficient, Big Data in Healthcare, Dynamic Data Distribution, Healthcare Information Systems, Partitioning Algorithms, Performance Evaluation in Databases
JOURNAL NAME:
Journal of Data Analysis and Information Processing,
Vol.13 No.1,
January
22,
2025
ABSTRACT: Efficient data management in healthcare is essential for providing timely and accurate patient care, yet traditional partitioning methods in relational databases often struggle with the high volume, heterogeneity, and regulatory complexity of healthcare data. This research introduces a tailored partitioning strategy leveraging the MD5 hashing algorithm to enhance data insertion, query performance, and load balancing in healthcare systems. By applying a consistent hash function to patient IDs, our approach achieves uniform distribution of records across partitions, optimizing retrieval paths and reducing access latency while ensuring data integrity and compliance. We evaluated the method through experiments focusing on partitioning efficiency, scalability, and fault tolerance. The partitioning efficiency analysis compared our MD5-based approach with standard round-robin methods, measuring insertion times, query latency, and data distribution balance. Scalability tests assessed system performance across increasing dataset sizes and varying partition counts, while fault tolerance experiments examined data integrity and retrieval performance under simulated partition failures. The experimental results demonstrate that the MD5-based partitioning strategy significantly reduces query retrieval times by optimizing data access patterns, achieving up to X% better performance compared to round-robin methods. It also scales effectively with larger datasets, maintaining low latency and ensuring robust resilience under failure scenarios. This novel approach offers a scalable, efficient, and fault-tolerant solution for healthcare systems, facilitating faster clinical decision-making and improved patient care in complex data environments.