Infobright-based data warehouse and system and its construction method

Abstract

A data warehouse based on Infobright includes a source data warehouse and a processing warehouse. The source data warehouse is used for storing raw data extracted from a relational database. The processing warehouse has a brighthouse engine, is in a cluster structure of one leader with multi-followers, and includes one processing main warehouse and a plurality of auxiliary warehouses. The processing warehouse is used for extracting the raw data from the source data warehouse, and processing and storing the raw data. The invention further provides a data warehouse system comprising the data warehouse based on Infobright, the relational database used for producing raw data relevant to business, and a data input unit used for inputting the raw data to the data warehouse. The invention further provides construction methods of the data warehouse and data warehouse system based on Infobright. The data warehouse has the advantages of high performances and scalability.

Description

technical field [0001] The present invention relates to digital data processing technology. Background technique [0002] Data Warehouse is a collection of data for analytical reporting and decision support for enterprises. The significance of building a data warehouse lies in: [0003] 1. Effectively integrate internal and external data of the enterprise to provide data basis for decision-making at all levels of the enterprise; [0004] 2. Integrate data from scattered and heterogeneous database systems; [0005] 3. Can analyze the profit and cost of the enterprise; [0006] 4. Standardize management, optimize process, and support decision-making. [0007] Common tools for building data warehouses include IBM Cognos and SAP B0. Both of the above adopt a centralized single-node mode to build a data warehouse, and this type of data warehouse has poor scalability. With the advent of the big data era, the scale of data has expanded rapidly, and some have reached the terab...

AI Technical Summary

Problems solved by technology

[0009] 1. Because the centralized single-node mode is adopted in the architecture, the horizontal expansion capability is poor, and it cannot be horizontally expanded in high-concurrency application scenarios;

[0010] 2. The data level support is limited, and it cannot support the data volume above TB level, so it cannot meet the massive data analysis of e-commerce, Internet and other enterprises;

[0011] 3. Because it is based on a traditional relational database, it is difficult to make a breakthrough in query performance, so that the user experience will be poor when the report is displayed;

[0012] 4. Expensive, which is not a small expense for enterprises

Images