The third article in the data warehouse series has finally started to talk about dimensions. Dimension, the word is very common, but also abstract. What is the dimension in the scenario of counting warehouses and scores? Check it out. 1. What is the dimension If you don't understand, ask, what is the dimension? Our natural response to learning is to consult professional materials. 1) Ali dataphin product introduction - the basic concept is to introduce dimensions as follows: the angle from which people observe things refers to a perspective , which is the multi-directional, multi-angle, multi-level conditions and concepts to determine things. 2) Huawei DGC product introduction - the basic concept is this way to introduce dimensions: dimensions are the perspectives used to observe and analyze business data , support data aggregation, drilling, and slice analysis, and are used for Group by conditions in SQL.
Most dimensions have a hierarchical structure, such as: geographic dimension, time dimension. 3) Look at what the "Data Warehouse Toolbox" has to say. Dimensions provide context around the " who, what, where, when, why, how " involved in a mobile number list business process. Dimension tables contain descriptive attributes needed by BI applications to filter and categorize facts. With a firm grasp of the granularity of the fact table, it is possible to distinguish all possible dimensions. When associated with a given fact table, the dimension should remain a single value in all cases. 4) Look at what "Alibaba's Big Data Road" has to say. Dimensions are the foundation and soul of dimensional modeling . In dimensional modeling, measures are referred to as "facts" and contexts are described as "dimensions," which are the diverse contexts needed to analyze a fact .
For example, when analyzing the transaction process, the environment in which the transaction occurs can be described through dimensions such as buyers, sellers, commodities, and time. The columns that a dimension contains that represent the dimension are called dimension attributes. Dimension properties are the basic source of query constraints, grouping, and report label generation, and are the key to data ease of use. Dimensions are generally used for query constraints, subtotals, and sorting . I don’t know how you felt when you first saw these explanations, but my real reaction was: the other party spoke to me well, why I couldn’t understand it. To be honest, these explanations are very professional and hard-core, but they are a bit difficult to understand. Partners who have not had much contact with data may be a little confused.