DRAG DROP
###BeginCaseStudy###
Case Study: 3
Data Architect
General Background
You are the data architect for a company that uses SQL Server 2012 Enterprise Edition. You
design data modeling and reporting solutions that are based on a sales data warehouse.
Background
The solutions will be deployed on the following servers:
• ServerA runs SQL Server Database Engine, ServerA is the data warehouse server.
• ServerB runs SQL Server Database Engine, SQL Server Analysis Services (SSAS) in
multidimensional mode, and SQL Server Integration Services (SSIS).
• ServerC runs SSAS in tabular mode, SQL Server Reporting Services (SSRS) running
in SharePoint mode, and Microsoft SharePoint 2010 Enterprise Edition with SP1.
The data warehouse schema currently contains the tables shown in the exhibit. (Click the
Exhibit button.)
Business Requirements
The reporting solution must address the requirements of the sales team, as follows:
• Team members must be able to view standard reports from SharePoint.
• Team members must be able to perform ad-hoc analysis by using Microsoft Power
View and Excel.
• Team members can have standard reports delivered to them on a schedule of their
choosing.
The standard reports
• Will use a sales territory hierarchy for organizing data by region.
• Will be accessible from SharePoint.
The Excel ad-hoc reports
• Will use the same data store as the standard reports.
• Will provide direct access to the data store for the sales team and a simplified view for
the executive team.
Technical Requirements
The standard reports must be based on an SSAS cube. The schema of the data warehouse on
ServerA must be able to support the ability to slice the fact data by the following dates:
• Order date (OrderDateKey)
• Due date (DueDateKey)
• Ship date (ShipDateKey)
Additions and modifications to the data warehouse schema must adhere to star schema design
principles to minimize maintenance and complexity.
The multidimensional and tabular models will be based on the data warehouse. The tabular
and multidimensional models will be created by using SQL Server Data Tools (SSDT). The
tabular project is named AdhocReports and the multidimensional project is named Standard
Reports.
The cube design in the Standard Reports project must define two measures for the unique
count of sales territories (SalesTerritoryKey) and products (ProductKey).
A deployment script that can be executed from a command-line utility must be created to
deploy the StandardReports project to ServerB.
The tabular model in the AdhocReports project must meet the following requirements:
• A hierarchy must be created that consists of the SalesTerritoryCountry and
SalesTerritoryRegion columns from the DimSalesTerritory table and the
EmployeeName column from the DimEmployee table.
• A key performance indicator (KPI) must be created that compares the total quantity
sold (OrderQuantity) to a threshold value of 1,000.
• A measure must be created to calculate day-over-day (DOD) sales by region based on
order date.
SSRS on ServerC must be configured to meet the following requirements:
• It must use a single data source for the standard reports.
• It must allow users to create their own standard report subscriptions.
• The sales team members must be limited to only viewing and subscribing to reports in
the Sales Reports library.
A week after the reporting solution was deployed to production, Marc, a salesperson,
indicated that he has never received reports for which he created an SSRS subscription. In
addition, Marc reports that he receives timeout errors when running some reports on demand.
###EndCaseStudy###
You need to complete the design of the data warehouse.
Which design should you use? (To answer, drag the appropriate tables and relationships to
the correct location in the answer area. Use only the tables and relationships that apply.)
http://examgod.com/bdimages/1bce4bff2b37_BF6F/wps18C3.tmp_thumb.png
Star Schema Design-> Dimension to facts
A – correct. We need to have star design. If you have a couple of fields with the same type you can bind it to multiple fields in fact table: https://www.simple-talk.com/sql/learn-sql-server/sql-server-data-warehouse-cribsheet/
B – also not bad, but too much work is done here
C – is ton a star
D – binds only to one attribute of fact table.
I think B, because must use tabular model.
A is better choice here as it’s meant to be the role playing Dimension in multidimensional project.
The multidimensional project is named “Standard Reports”.
The standard reports must be based on an SSAS cube. The schema of the data warehouse on ServerA must be able to support the ability to slice the fact data by the following dates:
• Order date (OrderDateKey)
• Due date (DueDateKey)
• Ship date (ShipDateKey)