Improving Clinical Data Warehouse Performance via a Windowing Data Structure Architecture

Data warehousing and online analytical processing (OLAP) are essential elements of decision support and commonly require large periods of time to process query results. Given that, it is crucial to tailor individual performance solutions based on the structure and merits of the data within the data warehouse to obtain the most optimal configuration. Therefore, this paper presents an analysis of the defining features of a clinical data warehouse, where the focus of the analysis is centered on the opportunities and threats for optimization purposes, based on the types of queries that are likely to arise. From the findings, the paper introduces a windowing data structure architecture (WDSA) to increase the performance of OLAP queries on a clinical data warehouse, which manages a collection of popular windows. The cost of processing a query using a simple instance of the WDSA was compared against existing query processing techniques such as nested join and hash join, to which the technique greatly outperformed both in almost all cases.