Performance analysis of Y-tree on flash drives

Abstract

Data warehousing is one of the most important fundamental components of business intelligence. One approach to improve a data warehouse is indexing. Y-tree is a type of index which is designed to facilitate one of the most critical data warehousing functionalities, i.e. data insertion. Y-trees support bulk insertion efficiently and also can provide the query performance as efficient as traditional value-list indexes. In this paper, we propose a theoretical study on the performance of the Y-tree indexes when implemented on flash drives. Our work will cover two main functionalities of Y-trees on flash drives, i.e. insertion and query performance. Given that flash drives are emerging as the main storage of the server-class computers, and their price per size are getting lower. The analysis of the performance in this paper can help to estimate the performance in practices. Also, the result can help practition to optimize Y-tree according to the given environment. © 2010 IEEE.