Uporaba počasi spreminjajočih dimenzij v podatkovnem skladišču

BOGATAJ, TINA (author), Zrnec, Aljaž (mentor), Poženel, Marko (comentor)

Pri načrtovanju podatkovnega skladišča se razvijalci pogosto srečajo z zahtevo po ohranjanju zgodovine sprememb. Za izpolnitev zahteve izberejo enega od osmih tipov počasi spreminjajočih dimenzij. Primernost tipa počasi spreminjajočih dimenzij se določi glede na potrebe poslovnih uporabnikov in števila podatkov, ki jih bodo hranili. V sklopu diplomske naloge bomo izdelali rešitev, ki bo uporabila eno od tipov počasi spreminjajočih dimenzij. Glavni del rešitve, ki bo reševal problem hranjenja zgodovine, bo razvit v sistemu Exasol s pomočjo skript, napisanih v programskem jeziku Lua. V rešitvi smo uporabili tip 7 počasi spreminjajočih dimenzij in s tem izpolnili zahtevo po hranjenju zgodovine sprememb. Z uporabo tega tipa lahko iz tabele dejstev neposredno dostopamo do podatkov, ki so bili veljavni ob vnosu zapisa, ali pa do podatkov, ki so veljavni sedaj.

• Language: Slovenian
• Keywords: podatkovno skladišče, počasi spreminjajoče dimenzije, Exasol, Airflow, PostgreSQL
• Work type: Bachelor thesis/paper
• Typology: 2.11 - Undergraduate Thesis
• Organization: FRI - Faculty of Computer and Information Science
• Year: 2023
• PID: 20.500.12556/RUL-144347
• COBISS.SI-ID: 142269955
• Publication date in RUL: 15.02.2023

Secondary language

• Language: English
• Title: Using slowly changing dimensions in a data warehouse

Abstract

When planning a data warehouse, developers are often faced with the requirement to maintain a history of changes. To meet the requirements, they choose one of eight slowly changing dimensions types. The type of slowly changing dimensions that suits the most business requirements is determined by the needs of business users and the amount of data they will store. As part of the thesis, we will create a solution that will use one of the types of slowly changing dimensions. The main part of the solution, which will solve the problem of preserving history, will be developed in the Exasol system with the help of scripts written in the Lua programming language. We used the slowly changing dimensions type 7 in the solution, thus fulfilling the requirement to keep a history of changes. Using this type, we can directly access the data that was valid at the time of entering the record or the data that is valid now from the fact table.

• Keywords: Data warehouse, slowly changing dimensions, Exasol, Airflow, PostgreSQL