当前位置:网站首页>Basic overview of data warehouse
Basic overview of data warehouse
2022-07-21 03:27:00 【Aisxi】
Data warehouse (data warehouse ,DW)
Definition
Data warehouse is a subject oriented integrated nonvolatile and time-varying data set .
It is mainly used to organize the accumulated historical data , And use analytical methods (OLAP, Data analysis ) Analyze and organize , And then assist in decision-making , For managers , Enterprises provide data support , Building business intelligence .
characteristic
1 subject-oriented : Gather the raw data according to the topic .
2 Integrate The original data source does not , Need to go through ETL conversion ( cleaning , Uniform standard )
3 Nonvolatile : Not allowed to be modified , Only query 、 analysis
4 Time varying : Several positions Will accept regularly . Integrate new data , So as to reflect the latest changes in data .
Several positions vs database
Database object-oriented design , Belong to OLTP System , The main operation is random reading and writing , Avoid redundancy in design , It is often designed with compound paradigm specification ;
Data warehouse is designed for theme , Belong to OLAP, The main operations are batch reading and writing , Focus on data integration , And analyze the processing performance , Will deliberately introduce redundancy , Design in an anti paradigm way ;
边栏推荐
猜你喜欢
随机推荐
【深度学习】怎么为模型选择正确的最后一层激活和损失函数
SQL子查询
Redis持久化
Opencv系列教程(一):Opencv读取指定文件夹图片、视频,调用摄像头
A. Log Chopping
Dameng DEM deployment
Spark SQL 中的小 TIPS
找一个数介于两个斐波那契数列之间
用户登录Demo
Redis主从复制
Redis五种基本数据类型(超级详细)
The role of 'defer' and 'async' attributes on the < srcipt> tag
Redis persistence
李宏毅2020机器学习--P17 CNN &P 14
Redis发布与订阅
OpenLayers Draw绘制时删除最后一个点
Chrome 进程架构
(三)PyTorch深度学习:反向传播梯度下降
达梦统计信息收集
(七)PyTorch深度学习:全连接层网络