The Distribution Algorithm of Data Flows Based on the BIRCH Clustering in the Internal Distribution Mechanism
Abstract
Today, the internal distribution approach is gaining ground in research related to large data streams. The internal distribution mechanism allows to achieve efficiency by applying the allocation mechanism on a single computing machine. In other words, storage, processing, and analysis of data flows based on the mechanism of internal distribution is one of the effective approaches in terms of both time and economic indicators. However, in this approach, the efficiency is closely related to the distribution module. That is, only the distribution of data flows based on specific rules will give the expected result. This research work is devoted to the development of an algorithm for the distribution of data streams in the internal distribution mechanism. In the study, the algorithm of distribution of data streams based on the Birch clustering method is considered. During the research, the capabilities of the clustering method are highlighted, and the distribution algorithm is developed based on this method. At the same time, the results obtained during the experiments are described and analyzed.