An Improved Query Technique for Embedded Database Based on S-B Tree

For the issue that embedded devices have limited space, this paper proposes a static balanced tree-S-B tree to substitute B⁺-tree as index of embedded database. Based on some embedded databases do not have to insert and delete on the actual application, S-B tree combines the brilliant time efficiency of B⁺-tree and the space utilization of static tree to reduce space waste by B⁺-tree index. We use this technique in the embedded database SQLite of Fixed Assets Management System. The results of experiment show, compared to B⁺-tree in the same conditions, the space utilization of S-B tree is above B⁺-tree about 30% on the average. S-B tree also has a better performance on sequence query and random query operations.