An Improved Algorithm of Binary Balanced Tree with Super Large-scale Data Set

Abstract

Realization of quick search still relies on ordered data in the application scenarios of large-scale data search and analysis. This paper analyses an improved algorithm of realizing super large-scale balanced tree efficiently in the concurrent environment. Size Balanced Tree, which is hard to build a balanced tree in the processing of mass data, can group the mass data and then combined groups into a SBT by multi path conflation algorithm. Experiment shows compared with other binary sorting trees(AVL, Treap, random BST), there is no detailed leap variation in time along with the increasing knots. This algorithm has a low time complexity and high running efficiency. It can be applied in the application scenarios of operating quick ordered search of super large-scale data in the artificial intelligence field.