A scheme of dynamic attributes addition for tuple datasets

Nowadays in response to the expansion and diversification of business operations, demands for dynamic change of dataset definitions (schema evolution) such as new attribute additions is increasing. However, in order to handle the tuples after the attribute additions efficiently, reorganization of the dataset is necessitated and high operation cost would be accompanied. History-pattern encoding scheme for dynamic multidimensional datasets is based on the notion of multidimensional extendible array. It can provide fast retrieval capability of the datasets while suppressing storage cost minimally. In this paper, using the history-pattern encoding scheme, we will presents the schemes of dynamic attribute addition for dynamically extending tuple datasets, and evaluate them.