Sample Modeling Questions
1) Look at DataSources and view the column InfoObject Template, what is the effect on that in generating automatic proposals?
Allows you to quickly create a referential structure based on existing InfoObjects. For example, we know that object 0ORDERDATE and the new object ZBIRTHDAY are the same date format so we can use 0ORDERDATE to quickly pull the correct data element information.
2) A Flat File name might change for the uploaded file from day to day. How can we ensure data will still load even though the file name changes?
Create a routine by filename. E.g. If the file is “sales_[date].csv” and the date changes every day when new data is loaded, use a wildcard in the routine to pick up “sales*.csv” files.
3) What is the exact structure of the PSA table?
Data packet number/Data record number followed by column headings and data for each column
4) If the same file is loaded 5 times with the same 5 records, you will not get a duplicate key error and 25 records in the PSA. Why?
The PSA organizes every load of data with a separate external id so nothing is mixed.
5) QUALOK, what is it, what are the different possibilities if not QUALOK, where is this found?
When you are loading data, the monitor for the request determines what actions can take place before, during, and after the load. When the load is green (complete), subsequent processing can and will begin. This is because the system knows the data is qualitatively correct. The default for QUALOK is all requests loaded successfully, meaning nothing is still loading and nothing has failed to load.
6) What is a process chain? Why are these used in SAP BW?
The process chain is the primary tool used to control and automate the ETL processes in BI. You use it to perform common, repeated tasks in your BI system that would otherwise require much manual effort. Companies use these to load data on a nightly basis and only intervene when required to. Think of it like an automated factory assembly line, sometimes the line jams, just like when bad data comes into SAP BW.
7) What is a DSO? Why do we need it? How many types are there?
What is DSO? Why do we need it? How many types are there?
A DataStore Object: Data in DSO objects is stored in flat, transparent, database tables integrating data that comes from different sources into one object. Multiple source consolidation at a detailed level data storage object would be too long to describe it, DSO works. We use DSOs for analysis and day-to-day decision making.
What are the three types of DSOs:
Standard DSO – consists of three tables (activation queue, active data table, and change log). It is completely integrated in the staging process. In other words, data can be loaded into and out of the DataStore Objects during the staging process. Using a change log means that all changes are also written and are available as delta uploads for connected data targets.
Write optimized DSO – It is targeted for the warehouse level of the architecture, and has the advantage of quicker loads.
Direct update DSO – Only has the active data table so not as easily integrated in the staging Process and relies heavily on BAPIs.
8) What is the old name for DSO?
9) What are the two major reasons we use DSOs?
Organization: Creates a consolidated dataset from one or several InfoSources
Analysis: Allows for up-to-date detail level analysis
10) T or F. You can load three different InfoObjects from the PSA at the same time.
True. The same data source may be used to load a SID table and a cube.
11) T or F. The same data source may be used to load a SID table and a cube.
12) T or F. If we are loading Transaction Data with all default options and the master data does not yet exist, the load will fail.
False. The load will succeed, however it will be empty.
Sample Reporting Questions
1) What is a key figure?
A key figure is a data field that stores values or quantities, such as amount. E.g. Units sold, revenue, quantity, costs, etc… can be key figures. Key figures are the facts in the fact table. There are 6 types: Amount, Quantity, Number, Integer, Date, and Time.
2) What is a characteristic?
In the SAP BW extended star schema the dimension attributes are used as descriptors of data or characteristics. They describe the objects used in business processes and are used to analyze key figures.
E.g. Vendors, customers, products which are all objects in a business processes. Other characteristics like weight and color, are attributes of characteristics that describe the data in even further detail.
Ex: If Vendor ID is a characteristic, its attribute can be vendor name and vendor address.
3) What is a restricted key figure?
A restricted key figure has been filtered or restricted, through the selection of one for more characteristics. They are used by BW queries to analyze specific data, such as quarterly sales. The KF to be restricted can be a basic KF, already restricted KF or a calculated KF. A restricted KF limits its restriction only to the selected KF’s characteristic value.
4) What is a calculated key figure?
A calculated KF is a value that is determined by using calculation rules or formulas within a report, such as average sales price. It can consist of basic, restricted, or other calculated KFs in an InfoProvider. The calculated KFs are stored in the metadata repository, not in DB tables as they are calculated in real-time.
5) In query designer, two types of restrictions can be set on the filter tab, what are they?
Character Restrictions and Default Values
6) You report on an InfoObject, but can’t find it in query designer, why?
If the InfoProvider is an InfoObject of the type characteristic, all of the characteristics listed in the attribute definition and all of the display attributes are assigned to the characteristic (and to the existing compounded characteristics) in the Key dimension:
· The display attributes are listed exclusively in the Key dimension.
· The independent characteristics are listed in both the Key and the Attribute dimensions.
If the InfoProvider of type DSO, a field object with the property “exclusive attribute” is not listed in the directory tree of the InfoProvider.
7) Use a cube only and report on all products that Company X did not sell. This is called a specific term, “slow moving”, is this type of reporting possible? Why?
Yes. We use slow moving reports when we want to display characteristic values for which there is no transactional data over a fixed time frame. A time range is required.
8) What is a display attribute?
Display attributes provide supplemental information when combined with a characteristic.
9) What is a navigational attribute? Are there prerequisites to use them in a report?
Navigational attributes can be used for report navigation. They enable navigation from characteristic to characteristic during query execution. A negative aspect of using navigation attributes is that they impact performance.
10) What is an error stack in your DTP?
Request-based tables (think PSA table) into which error records from a DTP are written.