BL的负担 (What a load of Bl…)
您可能想知道什么是商业智能。 如果您正在阅读本文，则您可能具有Microsoft SQL Server或至少一个其他关系数据库管理系统（RDBMS）的经验。 在这种情况下，您可能习惯于管理大量有价值的数据。 如果您是DBA，则可以通过重写存储过程，创建索引或运行性能分析跟踪来最大限度地提高性能。SQL Server Business Intelligence, on the other hand, is about making sense of these numbers and letters and transforming them into presentable information that provides decision-making value to the right people at the right time.
另一方面，SQL Server商业智能就是要理解这些数字和字母，并将它们转换为可显示的信息，从而在适当的时间为适当的人员提供决策价值。
Sounds easy enough right? Well, in reality, not really. It takes a mix of skills, tools, planning and a fair amount of time to put in place a large-scale BI solution to cover a modern company’s needs. Luckily, these requirements have led to the development of a large array of very handy tools which can be used to generate different types of reports in different formats.
听起来很简单吧？ 好吧，实际上并非如此。 部署大型BI解决方案以满足现代公司的需求需要技能，工具，计划和大量时间的综合。 幸运的是，这些要求导致开发了许多非常方便的工具，这些工具可用于生成不同格式的不同类型的报告。
Therefore, there are also ways of delivering rich business intelligence data to the end-user in relatively straightforward ways too.
因此，也存在以相对简单的方式将丰富的商业智能数据传递给最终用户的方法。
SQL Server商业智能的结构 (The structure of SQL Server Business Intelligence)
Central to all SQL Server Business Intelligence solutions exists a standardized method of preparing data to make it ready for consumption in the business intelligence world. This practice involves transforming data that is optimized for a heavily transactional system and moving it to one that is more suited for analytic querying.
所有SQL Server商业智能解决方案的核心都是一种标准化的数据准备方法，以使其准备在商业智能领域中消费。 这种实践涉及转换为一个事务量大的系统优化的数据，并将其移至更适合于分析查询的数据。
To understand this difference it is important to realize that a good production server is generally geared towards OLTP (online transaction processing) which means it has a highly relational structure, indexes and data types that are conducive to INSERTs, UPDATEs, and DELETEs. This is not the case further down the workflow towards business intelligence optimized data. Broadly, there are two types of systems with one being even more optimized than the other and therefore the endpoint in a mature BI dataflow.
要了解这种差异，重要的是要认识到，好的生产服务器通常适合OLTP（在线事务处理），这意味着它具有高度关系的结构，索引和数据类型，有利于INSERT，UPDATE和DELETE。 在朝着业务智能优化数据的工作流进一步发展的情况下，情况并非如此。 广义上讲，存在两种类型的系统，其中一种的优化程度甚至超过另一种，因此是成熟BI数据流中的终结点。
The first step away from an OLTP system is the Data Warehouse This is generally hosted in a normal SQL Server instance as a normal database however it has a de-normalized database model instead of a relational one. The most common structure is that star-schema model which has a central fact table containing measurable or calculable data surrounded by dimension tables that contain reference data that describe the fact, or measure, in the central table.
离开OLTP系统的第一步是数据仓库。数据仓库通常作为常规数据库托管在常规SQL Server实例中，但是它具有非规范化数据库模型而不是关系模型。 最常见的结构是星形模式模型，该模型具有一个中央事实表，该表包含可测量或可计算的数据，并由维表围绕，这些维表包含描述该中央表中的事实或度量的参考数据。
As mentioned previously it is possible to host both the OLTP instance and the data warehouse in the same SQL Server instance. It is also possible to start creating reports that query the data warehouse directly. It is therefore quite common for smaller companies to stop at this point. This is understandable as the less expensive license levels of SQL Server often do not include the Analysis Services tools required to move on the next stage of analytic processing.
如前所述，可以在同一SQL Server实例中同时托管OLTP实例和数据仓库。 也可以开始创建直接查询数据仓库的报告。 因此，较小的公司在这一点上停下来是很普遍的。 这是可以理解的，因为价格便宜SQL Server许可证级别通常不包括进行下一阶段分析处理所需的Analysis Services工具。
This next stage involves the creation of a Multidimensional Online Analytic Processing cube (MOLAP). It sounds like a mouthful and behind the scenes, it is quite complex but in practice, you can think of this very basically as a spreadsheet in three dimensions (hence the word cube). In reality, this cube can have many dimensions, but without having a profound knowledge of quantum physics the human mind is not really capable of imagining such a structure. In any case, the result is the same.
下一步涉及创建多维在线分析处理多维数据集（MOLAP）。 这听起来像是一张嘴，而且在幕后，它相当复杂，但是在实践中，您可以将其基本上看作是三维的电子表格（因此称为“立方体”）。 实际上，这个立方体可以具有许多维度，但是如果不具备量子物理学的丰富知识，人的大脑就无法真正想象这种结构。 无论如何，结果都是一样的。
You can pick a fact (or cell in your 3D spreadsheet) and drill down to a pre-calculated value along any chosen referential dimension.
您可以选择一个事实（或3D电子表格中的单元格），并沿任何选定的参考维度向下钻取到预先计算的值。
Putting all the jargon to the side, when you get to this stage you are able to query a read-only optimized database and return business intelligence rich results for extremely large datasets in blisteringly fast response times (when compared to similar queries on a traditional OLTP database).
把所有的术语放到一边，当您到达此阶段时，您可以查询快速优化的只读数据库，并以极快的响应时间返回非常大的数据集的商业智能丰富结果（与传统OLTP上的类似查询相比）数据库）。
Being multidimensional and read-only in nature a cube not only takes up less space on a disk but also avoids concurrency issues completely by not requiring locks.
由于多维数据集本质上是多维的并且是只读的，因此它不仅占用磁盘上较少的空间，而且不需要锁，从而完全避免了并发问题。
数据流 (The flow of data)
SQL Server Business Intelligence not only requires the underlying structure that supports these databases but also the ETL (Extract, Transform, Load) systems that allow the transfer of data between the OLTP database and the data warehouse. This is provided by SQL Server Data Tools (formerly known as SQL Server Integration Services or SSIS). This tool can be used for any operation that requires loading, transferring, enriching and sending data. It offers a bunch of handy tools to do anything from executing PowerShell files and running T-SQL scripts to more complex scripting in VB or C#.
SQL Server商业智能不仅需要支持这些数据库的基础结构，还需要允许在OLTP数据库和数据仓库之间传输数据的ETL（提取，转换，加载）系统。 这是由SQL Server数据工具（以前称为SQL Server Integration Services或SSIS）提供的。 该工具可用于需要加载，传输，充实和发送数据的任何操作。 它提供了许多方便的工具来执行从执行PowerShell文件和运行T-SQL脚本到VB或C＃中更复杂的脚本的所有操作。
The transfer between the data warehouse and the OLAP cube is dealt with in SQL Server Analysis Services (SSAS). This process entails mapping the data warehouse star-schemas to a cube structure by fact and dimension tables. This allows for the generation of an XMLA scripts which can construct the cube structure and process the transfer of data from the data warehouse to the cube.
数据仓库和OLAP多维数据集之间的传输在SQL Server Analysis Services（SSAS）中进行处理。 此过程需要通过事实和维度表将数据仓库的星型映射到多维数据集结构。 这样就可以生成XMLA脚本，该脚本可以构造多维数据集结构并处理从数据仓库到多维数据集的数据传输。
介绍 (Presentation)
Once the underlying structure is created and the flow of data between the different levels is scheduled the business intelligence developer must now find a way to present this data to the end-user.
一旦创建了基础结构并计划了不同级别之间的数据流，商业智能开发人员现在必须找到一种方法来将这些数据呈现给最终用户。
There are a number of ways to do this in SQL Server business intelligence. The most obvious and familiar being SQL Server Reporting Services (SSRS). This is a whole branch of SQL Server dedicated to building, hosting and presenting reports based on business intelligence data. Developers can create reports in Visual Studio and deploy them to the report server.
SQL Server商业智能中有许多方法可以做到这一点。 最明显和最熟悉的是SQL Server Reporting Services（SSRS）。 这是SQL Server的一个完整分支，致力于基于商业智能数据构建，托管和呈现报告。 开发人员可以在Visual Studio中创建报表，并将其部署到报表服务器。
With SQL Server 2012 we saw a greater integration with SharePoint and through Power View and Performance Point it is possible for end-users to create their own rich and dynamic reports through a web interface.
借助SQL Server 2012，我们看到了与SharePoint的更大集成，并通过Power View和Performance Point，最终用户可以通过Web界面创建自己的丰富而动态的报告。
One could also connection Excel spreadsheets directly to BI data sources of course. This is quite practical for the more adventurous end-user.
当然，也可以将Excel电子表格直接连接到BI数据源。 对于喜欢冒险的最终用户来说，这是相当实用的。
结论 (Conclusion)
SQL Server Business Intelligence is a whole other can of worms. There are patterns to follow and tools to use to create BI systems, transfer the data between them and there are other tools to create report hosting servers and the reports themselves. In the next article, we will take a closer look at each feature Now that you have understood the basics of SQL Server Business Intelligence it is time to get your hands dirty and learn about Data Tools, SSAS, SSRS, SharePoint and Power View.
SQL Server商业智能是蠕虫的另一种形式。 有一些可遵循的模式和一些工具可用于创建BI系统，在它们之间传输数据，还有其他工具可用于创建报表托管服务器和报表本身。 在下一篇文章中，我们将仔细研究每个功能既然您已经了解了SQL Server商业智能的基础知识，现在该动手学习有关数据工具，SSAS，SSRS，SharePoint和Power View的知识了。
资源： (Resources:)
Introduction to SQL Server Management Studio for Business Intelligence
Business Intelligence in SQL Server 2014 – SQL Server Data Tools (SSDT)
SQL Server Management Studio for Business Intelligence简介
SQL Server 2014中的商业智能– SQL Server数据工具（SSDT）
翻译自: https://www.sqlshack.com/sql-server-business-intelligence-introduction/