Topic Status: Some information in this topic is preview and subject to change in future releases. Preview information describes new features or changes to existing features in Microsoft SQL Server 2016 Community Technology Preview 2 (CTP2).
Analysis Services is an online analytical data engine used in decision support and business intelligence (BI) solutions, providing the analytical data for business reports and client applications such as Excel, Reporting Services reports, and other third-party BI tools. A typical workflow for Analysis Services includes building an OLAP or tabular data model, deploy the model as a database to an Analysis Services instance, process the database to load it with data, and then assign permissions to allow data access. When it's ready to go, this multi-purpose data model can be accessed by any client application supporting Analysis Services as a data source.
To create a model, use SQL Server Data Tools (see Tools and applications used in Analysis Services), choosing either a Tabular or Multidimensional and Data Mining project template. The project template contains folders for all of the objects needed in a model. You can use wizards to create all of the basic elements, such as data sources, data source views, dimensions, cubes, and roles.
Models are populated with data from external data systems, usually data warehouses hosted on a SQL Server or Oracle relational database engine (Tabular models support additional data source types). Models specify query objects, such as cubes, but also specify dimensions that can be used in multiple cubes, calculations and KPIs that encapsulate business logic, and interactions such as navigation and drill-through behaviors.
To use a model, it's deployed to an Analysis Services instance that runs databases in a particular server mode, making the data available to authorized users who connect through Excel or other applications.
As a Tabular instance, running Tabular models.
As a Multidimensional and Data Mining instance, running OLAP cubes and data mining models (this is the default).
As Power Pivot for SharePoint, running Power Pivot and Excel data models in SharePoint (Power Pivot for SharePoint is a middle-tier data engine that loads, queries, and refreshes data models hosted in SharePoint).
Same data engine; three different ways to use it. Note that server modes are set during installation and cannot be changed later. You should install a new instance if you require a different mode.
Foundational documentation for Analysis Services is organized into sections that correspond to the type of project you are building. Choose from the following links to learn more about each mode or feature area.