Microsoft Azure数据仓库:平面表或星型架构 [英] Microsoft Azure Data Warehouse: Flat Tables or Star Schema

问题描述

I am creating data warehouse model on numerous OLTP tables. a) I can either utilize a Star schema or b) Flat wide table model

许多人认为不需要维星昏暗事实模式模型；因为大多数数据都可以在一个表中报告.此外，当性能和存储成为问题时，将创建星型架构Kimball.一些人声称，随着技术的改进，数据可以 放在一个表中.

Many people think dimensional star dim-fact schema model is not required; because most data can report itself in a single table. Additionally, star schema Kimball was created when performance and storage are an issue. Some claim with improved tech, data can be presented in a single table.

我还是应该将数据划分为维度/事实表，还是仅在Azure数据仓库中直接使用平面表?

Should I still separate data into dimensions/facts tables or just use the flat tables directly in Azure data warehouse?

对于Redshift而言，许多论坛和员工都喜欢使用宽桌子. 

For Redshift, many forums and employees prefer flat wide table. Performance of Flat Tables Vs Dimension and Facts, 

在带有柱状存储的Microsoft Azure DW中，它们是平坦的宽表或星型图 推荐

In Microsoft Azure DW with columnar storage, are flat wide tables or star schema recommended?

推荐答案

BlueCar5，您好

Hi BlueCar5,

基于所提供的有限信息，散列表的散列可能是最好的选择:

Hashed distribution of large tables is probably the best option based upon the limited information provided: Hash distribute large tables. In many cases, any database is usable and meets the needs of the enterprise when initially set-up but after 6+ months of operations, things quickly deteriorate. So, how much data do you anticipate having to load over X period? What is the Data Warehouse being used for? My intent here is to flush out your requirements and get a dialog going to help you arrive at a working option or set of options. 

我在此处提供其他资源以供审核. 

I am providing additional resources here to review. 

Azure的最佳做法SQL数据仓库

SQL数据仓库容量限制

引用重复的堆栈溢出问题:

Referencing duplicate Stack Overflow question: Microsoft Azure Data Warehouse: Flat Tables or Star Schema

这篇关于Microsoft Azure数据仓库:平面表或星型架构的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！