INTELLIGENT ANALYSIS OF THE GREENPLUM DATA WAREHOUSE BASED ON LOG FILE PROCESSING

Rusakov Alexey Mikhailovich, Gorin Denis Stanislavovich, Lisyutenko Anastasia Sergeevna, Dolzhenkov Sergey Sergeevich, Karev Andrey Dmitrievich, Astafyev Ivan Andreevich

The article provides methods and algorithms for data mining contained in the debugging files of the GreenPlum data warehouse to improve the efficiency of the repository as a whole. One of the important tasks to be solved in this article is the detection of cause-and-effect relationships, as a result of which the GreenPlum storage is experiencing overloads. Based on the results of the study, a software package was designed and implemented in the form of an Internet portal that allows you to download the GreenPlum storage log files and find patterns in them that indicate shortcomings in the selected storage architecture with the possibility of automatic training based on existing precedents.

Keywords:Data storage, GreenPlum optimization, data mining, log file analysis