Analysis Services provides three
different approaches for creating a business intelligence semantic model:
tabular, multidimensional, and PowerPivot. Tabular solutions use relational
modeling constructs such as tables and relationships for modeling data, and the
in-memory analytics engine for storing and calculating data.
Multidimensional and data mining
solutions use OLAP modeling constructs (cubes and dimensions) and MOLAP, ROLAP,
or HOLAP storage. PowerPivot is a self-service BI solution that lets business
analysts build an analytical data model inside an Excel workbook using the
PowerPivot for Excel add-in. PowerPivot also uses in-memory analytics engine
within Excel and on SharePoint. Because PowerPivot solutions use Excel for both
data modeling and rendering, deploying a workbook on a server for centralized
and controlled data access requires SharePoint and Excel Services.
Tabular and multidimensional
solutions are built using SQL Server Data Tools and are intended for corporate
BI projects that run on a standalone Analysis Services instance. Both solutions
yield high performance analytical databases that integrate easily with Excel,
Reporting Services reports, and other BI applications from Microsoft and
third-party applications. Yet each solution differs in how they are created,
used, and deployed. This topic explores the differences, allowing you to
compare and identify the solution that best meets your project requirements.
Because tabular is the newer
solution, you might think that migrating an existing multidimensional solution
to a tabular format is the correct course of action, but this is usually not
the case. Tabular does not supersede multidimensional, and the two formats are
not interchangeable. Unless you have a specific reason to do so, do not rebuild
an existing multidimensional solution if it is already meeting the needs of
your organization. For new projects, consider the tabular approach. It will be
faster to design, test, and deploy; and it will work better with the latest
self-service BI applications from Microsoft.
The ability to define customized
management pack elements was used to model the online analytical processing
(OLAP) cube management pack elements that are included in
System Center 2012 – Service Manager. These management
pack elements make it possible for the user to declaratively define and
customize an OLAP cube at a higher level of abstraction.
Based on the definition, the
deployment of these management pack elements create the correct relationships,
components, and fundamental building blocks of the OLAP cube at a greater level
of detail, without any further user guidance. The following are the two main
management pack elements that are included in OLAP cubes:
- SystemCenterCube
- Cube Extension
No comments:
Post a Comment