Advanced Modelling

7/27/2019 Advanced Modelling

http://slidepdf.com/reader/full/advanced-modelling 1/116

MODELLING

&

ADVANCED MODELLING

Area:-

a folder where we can organize our project.

can create IA inside IA.

Object Catalog:-

a container 4 maintain more number of objects which are same type.have two types of catalogs

1)Characteristic.2)Key Figure.

1



ARACTERISTIC CATALOG

2



Y FIGURE CATALOG

3



4



Object:-

an entity for maintain the data.2 types

1)Characteristics: Object which is used to analyze data.

2)Key Figure: Object which is used to perform calculation.

racteristics:

5



6



7



Figure

8



9



10



11



12



13





es Of Characteristics:-

have 3 types of characteristics

1)UNIT CHARACTERISTC

It(Standard/Customized) maintains the units/currency

In case of customized it takes reference from 0unit/0currency

We can create unit char in 2 waysa)in T-code:RSD1

b)Create under not-assigned units catalog.

15



16





ME CHARACTERISTC

aintains the details of time like date, month, year, week, quarter . We can create time char in char as well as kf

ECHNICAL CHARACTERISTC

aintains all technical details likea)Request-id

b)Data Packet Number

c)Partition Number

d)Record Number

ter Data:-

ntains detailed information about the object.

ainly contains 3 tables.

1)Attribute Table.

2)Text Table.3)Hierarchy Table.

a)Text Table:It will helps u to maintain the object info in multiple LANGUAGES.It contains ATTRIBUTE KEY,LANGUAGE KEY,DESCRIPTION(SHORT,MEDIUM,LONG).

The naming convention of Text table is /BIC/T(INFO OBJ NAME).

18



b)Hierarchy Table:-

It maintains the parent-child relationship between diff attributes.If we r going 2 maintain the hierarchy on any obj that obj is in leaf level(lower level).

Highest level of node is parent node.

Max num of nodes are 98.

We have 3 types of Hierarchies1)Time Dependent(Structure depends on time).

2)Not-Time Dependent(Structure depends on time).3)Entirely Time Dependent(Entire Structure depends on time).

Select Type of hier

19









r activating hier we will get tree symbol under the char as follows.

23



24



c)Attribute Table:-

It maintains the detailed info about the object like cust num have name,adrs,…….etc.

If we are maintaining the attr of any obj that will act as key to it’s table.The naming convention of attr table is /BIC/P(OBJ NAME).

Attr always gives present truth only and they are over writable.

PES OF ATTRIBUTES:-have 7 types of attributes.

1)Display Attr.

It completely depends on Master characteristic in reporting. We can access in any Attr Table. It is available in

table. When ever we select ATTR ONLY that will b display attr.

25





eport

name,income will appear in result set when cno is in drill downy Figures

UST NUM UST NUM CUST'S NAME CUST'S INCOME Number of Records

00 VISH 100,000.00 1

01 SOM 200,000.00 102 CHAN 300,000.00 1

03 DAYA 400,000.00 1

rwise they won’t appear in result sety Figures

UST NUM

mber of Records

17

2)Navigational Attr

It is not depends on Master characteristic in reporting. We can access in any Attr Table. It is available in /BIC/

When ever we de-select ATTR ONLY that will be nav attr Where ever we want to use cno in any obj like(cusys will take nav_attr in nav part. And kf cann’t a nav attr .The naming convention of nav_at

M_CNO__M_CNAME(attr tab_attr name)”

elect attr-only

ch on nav_attr and select text to get description

will get nav_attr under obj

27





# # # 0.00 INR

Result 0.00 INR

Result 0.00 INR

erall Result 0.00 INR

y Figures ATE GION

NO

EGION STATE INCOME

# 0.00 INR

Result 0.00 INR

erall Result 0.00 INR

3)Time Dependent Attr

When we have an attr’s val is changing over period of time.And we want to maintain the historical dadepends/not depends on Master characteristic in reporting. We can access in any Attr Table. It is available in

table. When ever we select/de-select ATTR ONLY that will be TD attr.For this attr we will get 2 mo(DATE_TO , DATE_FROM)

We need to select Time-dependent

Table 4 time dependent attr

data

29



In reporting it shows diff data based on time

DDR NO NO ADDR

/C100 VIJ/C200 VIJN

/C300 BAN

/C400 DEL

DDR NO NO ADDR

/C100 VIJ/C200 VIJN

/C300 BAN1

/C400 NDEL

4)Time Dependent Navigational Attr

When we have an attr’s val is changing over period of time.And we want to maintain the historical da

depends/not depends on Master characteristic in reporting. We can access in any Attr Table. It is available in

table. When ever we select/de-select ATTR ONLY that will be TD attr.For this attr we will get 2 mo(DATE_TO , DATE_FROM).. It is not depends on Master characteristic in reporting. We can access in any Attr

When ever we de-select ATTR ONLY that will be nav attr Where ever we want to use cno in any obj like(cu

sys will take nav_attr in nav part. And kf cann’t a nav attr .The naming convention of nav_atM_CNO__M_CNAME(attr tab_attr name)”

We need to select Time-dependent Nav-attr.

Table 4 time dependent nav attr

30





NO STATE

/C100 AP1

/C200 AP4

/C300 KA2

/C400 DEL2

5)Exclusive Attr.

It is not depends on Master characteristic in reporting. We can access in any Attr Table oand obj like(cube/odavailable in /BIC/P table.When ever we de-select ATTR ONLY that will be exclusive attr.

NO y Figures

32



NO INCOME

/C100 0.00 INR

/C200 0.00 INR

/C300 0.00 INR

/C400 0.00 INR

6)Compound Attr.It is superior info obj. One value of info obj depends on value of other obj.It will act as other primary key to attr

Here cno is differed by plant number.

NO ANT

NO PLANT

/C100 P1

/C200 P1

/C100 P2

/C200 P2

7)Transitive Attr.

It is 2nd level of navigational attr. And it is Nav-attr of Nav-attr. It is not depends on Master charactereporting. We can access in any Attr Table. It is available in /BIC/X table. When ever we de-select ATTR ON

will be nav attr Where ever we want to use cno in any obj like(cube/ods) sys will take nav_attr in nav part.

cann’t a nav attr .The naming convention of nav_attr is ” M_CNO__M_CNAME(attr tab_attr name)”

33











2) If the relationship b/n 2 chars is M:1/1:M keep them in same dimension

3)If one char is attr of other char if clinent wants only present values then keep them in same dimension

4) If one char is attr of other char if clinent wants only fact values then keep them in diff dimensions

5)Reference characteristics should be in same dimension.

38



e need to set dimension as LINE_ITEM when we have only one char is assigned to it to increase reporting performwe have more data in cube we need to select CARD-HIEGHT to increase reporting performance

we place Key Figure in master table we will get only present or in fact table we will get only fact values

we have reports on month/year wise or quarter/year wise we need to perform partition on cube.ax num of sid are connected to dim table is 248 so total num of chars used in cube is 13X248=3224

ax num of kf connected to fact table is 233

PS TO DESIGN CUBE:entify the chars and kf.

Char kf

Cno qtyMno price

Date

entify info objects for each fieldNO,C_MNO,C_QTY,C_PRI,0CALDAY

aw BUBBLE diagram

aw the graphical star schema.

39



least we need to take 1 char,1 kf,1 time-char and we need to assign char 2 one dim. Other wise sys will show error

te cube under any info area.

40





GN THEM TO DIMENSIONS

42









ALLY WE WILL GET 18 TABLES FOR CUBE:

C/D(cube name)T -Dimension Table for TIME

C/D(cube name)U -Dimension Table for UNIT

C/D(cube name)P -Dimension Table for DATA PACKAGE

C/ D(cube name)1 -Dimension Table for (CUSTOMER GENERATED)

C/ D(cube name)2 -Dimension Table for (CUSTOMER GENERATED)

|

|

C/D(cube name)9 -Dimension Table for (CUSTOMER GENERATED)

C/D(cube name)A -Dimension Table for (CUSTOMER GENERATED)

|

46



|

C/D(cube name)D -Dimension Table for (CUSTOMER GENERATED)

C/F(cube name) -Fact Table

C/E(cube name) -Fact Table

PES OF CUBES:

1)Standard Cube(BASIC CUBE)

2)Transactional Cube

3)SAP Remote Cube

4)General Remote Cube.5)With Services Cube.

dard Cube:

ECT CHARACTERISTICS

47



GN THEM TO DIMENSIONS

48













we have (BASIC CUBE + TRANSACTIONAL)properties.

his cube we have 2 types of modes for loading data

1)Loading Mode(Normal ETL process)2)Planning Mode(User can enter data manually)

can put any mode but we cann’t allow 2 modes at a time(We can select any mode and load data)

planned mode we can load data with the T-Code : RSINPUT

r the data manually

54











primary key must be there.

can use data elements too.

59





61





eate Data Source For Table.to RSO2.

can create DS .

63



eate Info source wit MYSELF/ R/3.

64



eate info cube 4 SAP Remote Cube

65





67





lendar Year urce System y Figures

UST NUM AT NUM UST NUM MAT NUM Source System Calendar Year QUANTITY PRICE

M1 My IDES BW & SEM 2009 123 EA 1,230.00 INR

M2 My IDES BW & SEM 2010 123 L 1,230.00 AUD

M1 My IDES BW & SEM 2011 123 KG $ 1,230.00erall Result 369.000 MIX 3,690.00 MIX

eral Remote Cube:

Virtual Provider.

cess the source system when we want to fetch the data remotely from NON-SAP application at the run time.

on’t store data physically.tracts the data 4m remote NON-SAP appl and gives the data 2 report.

eate Table in NON-SAP application server epare Table in SQL server.

****** TO CREATING TABLE.

> CREATE TABLE TRAN(CNO VARCHAR(5),MNO VARCHAR(5),ZDATE NUMBER(8),QTY NUMBER(5

RCHAR(5),NUMBER(5),CURR VARCHAR(5));

e created.

69





71





eate Cube with Generic Remote Cube

73







Reporting

76



S(OPERATIONAL DATA STORE):

2 dimensional table.

er writes the data records with same Key Field combination.mainly used for PERMINENT staging purpose(PSA 4 TEMPORARY staging)

aintains the GRANULAR DATA.

have 2 main parts in ODS

1)KEY Fields(acts as keys 4 ods max num of key fields are 16 only characteristics but not kf)2)DATA Fields(except kf all objs are df only max num of df are 749)

have 3 tables in ODS

1)NEW Table2)Change Log Table

3)Active Table

n ever we load data into ODS data will loaded into NEW Table the req will be in yellow color. and we need to n color and activate request when ever we activate request the data will goes into active table. while moving data 4

to active table the change log stores the changes in recs which contains same kf combination

77







NALLY WE WILL GET 3 TABLES FOR ODS:

C/A(ODS Name)40 -NEW Table

C/A(ODS Name)00 -ACTIVE Table

C/B000*ODS-NAME* -CHANGE LOG Table

W Table-(KEYS{LOADING REQ+DATAPACKID+RECORD}+KF+DF+RECORD MODE)

80



TIVE Table-(KEYS{KF}+DF_RECORD MODE)

ANGE LOG Table-(KEYS{LOADING REQ+DATAPACKID+RECORD+PART NUM}+KF+DF+RECORD

ord Mode -N-new img

[ ]-after imgx-before img

81





PES OF ODS:

1)Standard ODS(BASIC ODS)

2)Transactional ODS

dard ODS:

r above ods

Reporting

lendar Year

83





ant to load data thorough RSINPUT

85





reporting create INFOSET and generate report on tran-ods

lendar Year

UST NUM

AT NUM y Figures

AT NUM QUANTITY PRICE

87









91















O SET:

mbines only ODS and INFO OBJECT.

Virtual Provider.

cess the all info providers under INFO SET when we want to extract the data at the run time.on’t store data physically.

orks with JOIN.

cann’t access Any type CUBEnot a view

ESIGNING MP WE HAVE ATLEAST ONE CHAR MUST BE COMMON-data element.

have 3 types of joins

1)Inner Join(ODS-ODS && ODS-INFO OBJ)It Brings only combined data

2)Outer Join(ODS-ODS && ODS-INFO OBJ)

It Brings combined data as well as extra records 4m opposite ods3)Temporal Join(ODS-ODS && ODS-INFO OBJ)

It Brings the data related to date

Ex:-if we have month in one ODS and date, reg in ODS1. if we put TEMP JOIN and if we give month

sys will shows reg at particular mont

te 1st ODS 4 cust details:

98



te 2nd ODS 4 Transactional data

99



can create mp under any IA.

ner Join

100





01 SOM SOM 2 KG $ 20,000.00

02 CHAND CHAN 3 L 50,000.00 AUD

03 DAYA DAYA 1 EA 15,000.00 INR

04 MANJ MANJ 2 KG $ 20,000.00

05 NAG NAG 3 L 50,000.00 AUD

06 RAJU RAJU 1 EA 15,000.00 INR

07 SESHU SESH 2 KG $ 20,000.00

08 LAX LAX 3 L 50,000.00 AUD

09 KITTU KITT 1 EA 15,000.00 INR

10 DEV DEV 2 KG $ 20,000.00

11 PHANI PHAN 3 L 50,000.00 AUD

12 RAMU RAMU 1 EA 15,000.00 INR

13 BALU BALU 2 KG $ 20,000.00

erall Result 27.000 MIX 375,000.00 MIX

EFT OUTTER JOIN

UST NUM 4 DET

lendar Year

AT NUM

y Figures UST NUM 4 TRAN UST'S NAME

102



UST NUM 4 TRAN CUST'S NAME QUANTITY PRICE Number of Records

00 VISH 1 EA 15,000.00 INR 1

01 SOM 2 KG $ 20,000.00 1

02 CHAN 3 L 50,000.00 AUD 1

03 DAYA 1 EA 15,000.00 INR 1

04 MANJ 2 KG $ 20,000.00 1

05 NAG 3 L 50,000.00 AUD 1

06 RAJU 1 EA 15,000.00 INR 1

07 SESH 2 KG $ 20,000.00 1

08 LAX 3 L 50,000.00 AUD 1

09 KITT 1 EA 15,000.00 INR 1

10 DEV 2 KG $ 20,000.00 1

11 PHAN 3 L 50,000.00 AUD 1

12 RAMU 1 EA 15,000.00 INR 1

13 BALU 2 KG $ 20,000.00 1

RANG 0.000 0.00 1

RATN 0.000 0.00 1

D OTHER INFOSET 4 OUTTEER JOIN

UST NUM 4 DET lendar Year

AT NUM

y Figures UST NUM 4 TRAN UST'S NAME UST NUM 4 TRAN CUST NUM 4 DET CUST'S NAME QUANTITY PRICE N

00 C100 VISH 1 EA 15,000.00 INR

01 C101 SOM 2 KG $ 20,000.0002 C102 CHAN 3 L 50,000.00 AUD

03 C103 DAYA 1 EA 15,000.00 INR

04 C104 MANJ 2 KG $ 20,000.00

05 C105 NAG 3 L 50,000.00 AUD

06 C106 RAJU 1 EA 15,000.00 INR

07 C107 SESH 2 KG $ 20,000.00

08 C108 LAX 3 L 50,000.00 AUD

09 C109 KITT 1 EA 15,000.00 INR

103



10 C110 DEV 2 KG $ 20,000.00

11 C111 PHAN 3 L 50,000.00 AUD

12 C112 RAMU 1 EA 15,000.00 INR

13 C113 BALU 2 KG $ 20,000.00

14 # # 3 L 50,000.00 AUD

15 # # 1 EA 15,000.00 INR

16 # # 2 KG $ 20,000.00

EMPORAL JOIN

104



N-CUMULATED KEY FIGURE:

n we don’t want the kf values to be cumulated we go 4 Non-Cumulated Key Figure.sing Exception Aggregation we will define nckf.

rio1:-According 2 year we want number of Employee

te Char,Kf

105



te cube and load data

eporting we will get like.lendar y Figures

OMPANEY OMPANEY Calendar Year NUMBER OF EMPLOYEES

2009 2,500

2010 2,000

2011 3,000erall

lt3,000

106





rio3:-Calculate one emp how many times visit each shopte Char,Kf

lendar Day

y Figures MP

OP MP SHOP AMOUNT

00 S1 1

S2 1

S3 1

S4 1

Result 4

01 S1 1

S2 1

S3 1Result 3

EN HUB SERVICES:

OSPOKE:

etract the data 4m objects into files in external system’s file and table

can retract data data into file in Appl-Server, Client Work Station, Data Base Table.

108



n ever we are trying 2 retract data 2 ff inas,cw we will get 2 files they are1)(Info spoke name).CSV

2)S_(Info spoke name).CSV

109



Appl-Server

110



111



112





poke;INFO_SPOKE1

Stamp;20020120183821

r\sap\SBW\DVEBMGS01\work\INFO_SPOK.CSV

UMN;FIELDNAME;KEY;TYPE;LENGTH;OUTPUTLEN;DECIMALS

1;CALDAY;X;DATS;000008;000010;000000

2;CALMONTH;X;NUMC;000006;000007;000000

3;CALMONTH2;X;NUMC;000002;000002;000000

4;CALQUART1;X;NUMC;000001;000001;000000

5;CALQUARTER;X;NUMC;000005;000006;000000

6;CALWEEK;X;NUMC;000006;000007;000000

7;CALYEAR;X;NUMC;000004;000004;000000

8;CHNGID;X;NUMC;000014;000014;000000

9;CURRENCY;X;CUKY;000005;000005;000000

10;FISCPER;X;NUMC;000007;000008;000000

0000;000000;00;0;00000;000000;0000;00000000000000;AUD;0000000;K1;0;REQU_A9EPGXEP7O682809LQQUCCRHH;K0

0000;000000;00;0;00000;000000;0000;00000000000000;INR;0000000;K1;0;REQU_A9EPGXEP7O682809LQQUCCRHH;EA

Client Work Station

114



Data Base Table

e select full update we need to select this for reducing duplication

e select this we will get 3 extra technical fields they are REQID,DPKTID,RECNO

115



Advanced Modelling

Documents