Granular Division and Calculation Process of Pyramidal Algorithm Based on Massive Data

Abstract

With the continuous application and operation of the software system for many years, the amount of data accumulated in the system database will be larger and larger, resulting in a slower and slower calculation of seemingly simple statistics such as sum and average, which seriously affects the stable operation and user experience of the system. According to pyramidal algorithm, this paper first define and use the optimal accumulative total edge to realize the process of rapid accumulation, and then put forward the granularity classification of the algorithm, positive cumulative, reverse the accumulate and mixed, and presents the automatic selection for statistical methods according to actual condition. Finally, the paper gives the application process and operation effect of the method, the results show that the granularity division of pyramidal algorithm provides the basis for the partition and decomposition of massive data.