从哪里开始多维数据仓库（SQL Server 2008 & VS2010）
我对这个概念很陌生......
我对数据仓库有一些理论知识，但我不知道从哪里开始多维数据仓库......
所以请给出一些关于从
- 哪里开始
- 的建议（工具或东西） ）适合多维数据仓库。
在我的系统中，我有
- SQL Server 2008(SQLManagementStudio_x86_ENU.exe,SQLEXPR_x86_ENU.exe)
- Visual Studio 2010(10.0)
I am new to this concept...
I have some theoretical knowledge on data warehousing, but I don't know where to start the multidimensional data warehousing...
So please give some suggestions regarding about
- where to start
- which is (tool or stuff) suitable for multidimensional data warehousing.
In my system I have
- SQL Server 2008(SQLManagementStudio_x86_ENU.exe,SQLEXPR_x86_ENU.exe)
- Visual Studio 2010(10.0)
如果你对这篇内容有疑问，欢迎到本站社区发帖提问 参与讨论，获取更多帮助，或者扫码二维码加入 Web 技术交流群。
绑定邮箱获取回复消息
由于您还没有绑定你的真实邮箱，如果其他用户或者作者回复了您的评论，将不能在第一时间通知您！
发布评论
评论（2）
首先要意识到数据仓库是商业智能解决方案的组成部分。其他组件包括表示层和集成层。在开始构建解决方案的一部分之前，您应该考虑整个解决方案。
接下来，“多维”是什么意思？不同的人可能由此理解不同的事情。这可能意味着您想要支持像 Analysis Services 这样具有自己的“多维数据集”模型的工具。或者这可能意味着您正在考虑为您的数据仓库采用 Kimball 式的“维度”设计。
当您觉得自己有足够的信息来开始设计实际的仓库数据库时，您可以像任何其他数据库设计一样开始。识别属性、键和依赖关系，并设计至少符合 Boyce-Codd / 第五范式的数据模型。如果你以前没有这样做过，我建议你先读一本好书。您还需要考虑诸如一致性不同来源的数据、分配代理键和对更改数据进行建模等注意事项。
Firstly be aware that a Data Warehouse is one component of a Business Intelligence solution. Other components include the presentation layer and integration layers. You should be thinking about the whole solution before you begin building part of it.
Next, what do you mean by "multidimensional"? Different people may understand different things from this. It could mean you want to support a tool like Analysis Services that have their own "cube" model. Or it could mean you are considering a Kimball-style "dimensional" design for your data warehouse.
When you feel you have enough information to start designing the actual warehouse database, you can start much as you would any other database design. Identify attributes, keys and dependencies and design a data model to at least Boyce-Codd / 5th Normal Form. I suggest you study a good book first if you haven't done this before. You also need to take into account considerations like conforming data from different sources, assigning surrogate keys and modelling changing data.
如果您掌握了所有知识，从哪里开始？
最合适的工具取决于。
您很可能拥有 2 个数据库和一组多维数据集 - 大多数人/项目将 Staging 和最终数据库模型分开。
Where to start, if you have all the knowledge?
Tools most suitable depend.
You most likely will have 2 databases and a set of cubes - most people / projects split off Staging and the final database model.