1. What types of extractor are available?

2. How to create Generic datasource?

Go to T. Code RSO2

3. What options or types available to create generic data source?
3 options available. (i) Extraction from DB view or table (ii) Infoset query (SAP Query)
(iii) Extraction from Function Module
4. What is generic delta?
To convert generic data source delta capable we use generic delta options.

5. What options available to create generic data source as delta capable?
6. What are delta specific fields in generic datasource?
To convert generic data source as delta capable can use generic delta options
We are having 3 types of delta specific fields
(i) Timestamp
(ii) Calendar day
(iii) Numeric pointer


7. What is safety intervals in generic delta and what types available?
Safety intervals are used to reduce chances to miss delta from generic data sources.
There are two types of Safety intervals available.
Upper Safety Interval
Lower Safety Interval
Let understand upper safety interval with use of timestamp field. When we choose upper safety interval option
then current highest value of delta will reduced with given safety interval. Assume we have given upper safety
interval as 300 seconds and data load to SAP BI occurs every one hour and started at 12:00, then data load will
be like that



Means in upper safety intervals it will leave last 300 seconds / 5 minutes so chances of missing current 5 minutes
records which may be is getting edited will be reduced. And next delta load will start from 11:55 onwards to current
time minus 5 minutes
In Lower safety Interval this 5 minutes data will load twice to SAP BI but in this case we cannot create Load this data
directly in Infocube otherwise we will have data duplication error.
8. What is New status for changed records and additive delta in generic delta?

9. How you will decide to create Infocube?
First I have to understand what type of requirement is this. Means Analytical or operational reporting
12:00 13:00 14:00 15:00
11:55 12:55 13:55 14:55
Then I will see which level of details they are looking for reports, if aggregated then probably Infocube is best option.
Then I will Identity the relationship and cardinality between characteristics
1:N cardinality related characteristics I will keep in same dimensions
N:M cardinality related characteristics I will keep in different dimensions
10. How much dimensions we can create in Infocube and why only ?
We can create max 16 dimensions in which 3 are reserved by SAP. 16 dimensions limit came from database
limitations, as in a table we cant have more than 16 keys as primary key.
11. What is Line Item Dimension?
When Dimension size is more than 20% compare to Fact table size then we need to optimize the Infocube
modelling.
Line Item Dimension (LID) is a dimension in which only one characteristic can be kept. When mark any dimension as
LID then Dimension table will dropped and directly SID will get inserted in Fact table.
Pros: 1.) Faster read access as one table got removed so no need to look into dimension table
2.) Faster data load as no need to generate Dimension ID
Cons: Only one characteristic can be keep in this dimension.
Eg.: Sales order no, Document No, Billing no type of characteristics should me make as LID, as mostly will
have 1:1 relationship with Fact table.
12. What is High Cardinality in Infocube?
When Dimension size is more than 20% compare to Fact table size then we need to optimize the Infocube
modelling.
When we are having more than one characteristics in dimension which increases dimension table size more than
20% compare to fact table. Then this type of dimensions should mark as High Cardinality Dimension.
Internally in database instead of BITMAP index BINARY Tree index will generate for this dimension, It will help to
faster load the data to Infocube. So by marking HC only data load performance will get improve. Thats why we have
to make sure this dimensions characteristics should not use frequently in reporting.

13. How many fact tables are available in Infocube?
We are having 2 fact tables in Infocube. F-Fact table and E-Fact table
F-fact table is partitioned with respect to Request ID.
E-fact table is compressed fact table, here no request id will available. Data will move from F-fact to E-fact table
while compression
14. What are aggregates?
Aggregates are the baby infocubes or we can say aggregates are the materialized subset of Infocube. As per
reporting need we can create N-No of aggregate on a Infocube. Aggr. Can be created based on characteristics, nav.
Attr., Time depended Nav. Attr., Hierarchy and Hierarchy node. While roll-up new request data of infocube will add
to aggregate.
15. How many types of DSOs are available?
There are 3 types of DSOs are available.
(i) Standard DSO
(ii) Write Optimized DSO (WDSO)
(iii) Direct update DSO

16. What is the star schema, classical star schema, snow flake star schema and extended star schema ?
Refer Wikipedia and internet
17. What is normalization? Also describe 1
st
form, 2
nd
form, 3
rd
form, 4
th
form and 5
th
form of normalization?
Refer Internet
18. What are the steps involved in LO data extraction?
a. RSA5 Select the Datasource and activate
b. LBWE Maintain Datasource add fields from communication structure if you need any then Activate
Extract Structures
c. Select appropriate update mode
d. LBWG Delete Setup Tables
e. Fill setup table (0LI*BW) or from menu path (SBIW)
f. RSA3 Check extraction and the data in Setup tables
g. Replicate data source in BI
h. LBWQ to check the extraction queue
i. RSA7 BW Delta Queue Monitor
19. Most generally used T. Codes used in LO data extraction?
LBWE LO cockpit
LBWG delete setup table
LBWQ Extraction Queue
RSA7 Delta Queue
SM13 Update tables
RSA5 Business content Data source repository
OLI*BW fill setup table
SBIW Menu path Navigation

20. Difference between ODS and Infocube
ODS Infocube
Provides Granular/detailed data Provides Aggregated data
Works on overwrite property Works on additive property
Follow 2 dimension modeling follow multi dimension modeling
We cannot create aggregate on
DSO
We can create aggregate on DSO
for Query optimization

21. How many types of routines are available in transformation?
We are having four types of routine available in transformation.
(i) Start Routine: It executes before the transformation. It is having structure or internal table as source
structure. We mostly use this for data selection or data filtration. It executes ones per data package.
(ii) End Routine: It executes at last before storing data to target. It is having structure or internal table as
target structure. If we want to do some calculation before storing data to data target then we will use this. It
executes ones per data package.
(iii) Field or Characteristics routine: This routine is specific to particular field/char. If we want to do
calculation for particular field we will use this routine. This routine will get execute for every record.
(iv) Expert routine: When we are not have any field to map with transformation and every field of data
target is getting calculated via code then we probably choose this routine. Here everything we need to write
with code. Expert routine having source structure, target structure both. When we write expert routine
there is no transformation at all and we cannot use any other routine with expert routine.








Source Target
S
T
A
R
T
E
N
D
Fiel
d
Fiel
d
Fiel
d
22. What types of data available in SAP server? (Classes of data)
There are three types of data available in SAP server.
(i) Master data
(ii) Transaction data
(iii) Configuration data : Its system sitting data to manage server. Mostly Basis team deal with this data.
(i) Master Data is further classified into 3 types:
a) Attribute Data: It describes things
b) Text Data:
c) Hierarchical Data:
(ii) Transaction Data is further divided into 2 types:
a) Document Data
i. Header Data
ii. Item Data
iii. Schedule line Data
b) Summary Level Data
23.