SAP Bo Technology

1. connection path problems:

-Loop

In relational database mode, a common join path that returns too few rows is called a loop. A loop is a group of joins. It defines closed joins of a group of tables in the pattern, for example:

Solution: Create an alias table (a copy of the base table ).

-Chasmtrap)

The fault trap is a common problem in the relational database model. The connection path returns more data than expected, A fault trap is a join path between three tables when two "many-to-one" joins in one table,

The context in the converged connection path is not properly separated, for example:

Solution: create a context for each fact table. This method applies to all situations.

-Fantrap)

Slice traps are common issues behind the fault traps in relational database models. Their impact is the same as that of returning more data than expected, when the "one-to-multiple" join link links tables connected by another "One-to-multiple" in turn,

A sector trap is a join path between three tables, such:

Solution: Create an alias table (a copy of the base table), create a new join between the alias table and the base table, and finally build the context.

 

2. What are the concepts, implementation methods, and advantages and disadvantages of derived tables and alias tables?

The derived table is similar to the database view, and the alias table is renamed in the universe of the database physical table;

The derived table can flexibly process complicated data logic, but it is inefficient;

The alias table is set based on business needs and is generally used to solve the loop problem in universe.

 

3. Key Points of universe design and precautions?

Design Points:

A. Analysis and Planning of business problems universe Solutions

B. Design Model

C. Construct universe

D. Distribute universe to users

E. Dimension objects are placed separately by Dimension

F. Place measurement objects by topic

G. The prompt corresponds to the dimension and topic.

Note: context, loop, fault trap, slice trap, base, and Bo development specifications.

 

4. How to quickly analyze, locate, and solve Bo report problems (TEST ):

 

5. Report performance optimization policies:

A. logical operations move to the backend, that is, the report layer moves to the Universe layer and the Universe layer moves back.

B. Add query conditions on the report to reduce the data volume returned for analysis.

C. Create a Materialized View

D. Avoid creating complex universe. A universe contains many fact tables and can be split as appropriate.

E. set multiple data requests for Parallel Data Query

F. Database optimization (Table Indexing and association conditions)

 

6. custom development:

The goal is to make it easy for customers to use. objects such as measurement, latitude, and filtering conditions should be made public and merged as much as possible, and named as easy-to-understand names, so that users can drag and drop combinations by themselves.

 

7. Variable creation:

Both universe and webi reports can create variables, but universe variables can be used in all reports of the Universe, while the variables of the webi report can only be used by the report itself.

 

8. Merge query, alarm, and dimension of webi reports:

Merge query is the union operation in SQL. Each query must maintain the consistency of the number and type of query objects. Alarms are used for data warning. When the data meets certain conditions, specific content can be displayed as needed. Merge dimensions merge the same dimensions in multiple queries into one latitude. When a merge dimension is used in a report, the report searches for the latitude data of the corresponding query based on the metric. However, if the report is complex, the merged dimension function may be disabled.

 

9. Under what conditions should I use multiple queries?

A. Different query themes

B. Low query efficiency

 

10. There is no problem when the SQL statement automatically generated by the report is obtained from the database, but the report result is incorrect because:

There is a problem with the database connection of universe.

 

11. Context understanding:

A context is a set of joined sets, that is, the association relationship between a group of database tables. It provides a valid query path for the webi report so that the report can generate correct SQL statements.

 

12. Test: first, the Bo consultant should perform the unit test on his own. The Bo consultant must first ensure that the variables and logic made by Bo are correct. Generally, data problems are mostly BW problems. Second, bo consultant and BW

The consultant performs integration testing together. Finally, the customer will be asked to perform the test after there are no problems in both aspects.

 

13. For the Bo project with BW as the data source, the consultant focuses on the development report. Universe is automatically generated, so it does not need to spend too much work; before development, you must specify the report format and the type of Bo report tool used with the customer.

The surrounding system (non-SAP system) is a data source Bo project, focusing on the universe design, first designing a correct and reasonable universe, and then designing a report.

SAP Bo Technology