Key less Technique in Data Stage

Key less Technique

SAME To more previously staged partitioning to next corresponding stage with    same  key column values. ENTIRE Can only be used in  lookup stage with reference  link when it is “normal lookup”   In Entire what ever the records that are processing on N1, the same set of records are processed on the rest of the node.

At TekSlate, we offer resources that help you in learning various IT courses.
We avail both written material and demo video tutorials.
To gain in-depth knowledge and be on par with  practical experience, then explore DataStage Training Videos.

RANDOM - Default / Auto Round Robin Generates the next number with sequence Note :- All the Generating stages gives the best performance when they are associated with Round Robin.

Generating  Stages

column generator , Row generator, Surrogate Key stage RANGE  It is an Expensive partitioning Technique than  Mod and Hash . This technique  needs data pre processing. Note :- Stage  :- Write Range Map (Development/ Debug stage)  To create the range if we want to use Range partitioning Web Console log in to “IBM information Server” with Same designer ID and PasswordàAdministration users and groups users user open

• To create New user  New
• Do main Management  license up gradations
• Session Management login time
• Users and groups
• Log Management  Access all logic from Director
• Scheduling Management Scheduling vices

Reports We  can view and generate reports Web console Admin DS AdminProject level Administration  (or)  Repository level

Brief on Surrogate key in DataStage

Surrogate key

• A kind of Alternative primary key
• Generate sequence number

Output

Sid	Cid	CN	Add
1	11	A	HYD
2	22	B	SEC
3	33	C	Del
4	22	B	GNT
5	44	E	PUN

If we don’t take surrogate key it maintains a new column active flag(indicate current address) o/p:

CID	CN	ADD	A.F
11	A	Hyd	Y
22	B	SEC	Y/N
33	C	Del	Y
22	B	GNT	Y
44	E	PUN	Y

Ex: After data;(source)

CID	CN	ADD
11	A	HYD
22	B	Cal
66	G	ASm

Before Data(Target)

CID	CN	ADD	A.F
11	A	Hyd	Y
22	B	SEC	Y
33	C	Del	Y

Output Before data(Target)

CID	CN	ADD	A.F
11	A	Hyd	Y
22	B	SEC	Y/N
33	C	Del	Y
22	B	GNT	Y
66	G	ASM	Y

After Data: (Source)

CID	CN	ADD
11	A	GNT
22	B	DEL
55	H	MP

Output

CID	CN	Add	A.F	Effective start date	Effective end date
11	A	Hyd	Y/N	3-6-06	(9999-12-31) 26-10-10
22	B	SEC	Y/N	3-6-06	(99-12-31) 27-9-07
33	C	Del	Y	3-6-06	99-12-31
22	B	Cal	Y/N	28-9-07	(99-12-31) 26-10-10
6	G	ASm	Y	28-9-07	99-12-31
11	A	GNT	Y	27-10-10	99-12-31
22	B	DEL	Y	27-10-10	99-12-31
55	H	MP	Y	27-10-10	99-12-31

 Active Tag :- (A.F) New  records will maintain flag ‘Y’ Old records ‘N’ Tables will data maintain effective start data and effective end data If data is not displayed, then we get a new column record version , unique id

Record version, unique id SCD  - Type  Table 1 Create table SRC   (SNO number (3), S name   Varchar 2 (25)); Create the table and reject values Table 2 Create table CIM   (SNO number (2), S name   Varchar 3 Sname varchar2 (25)); Just create the table only 5 steps for SCD – Type 1 

S1:- Select output link as Fact link

S2: Map the lookup key SNO

S3: Source type  = flat file

= Source Name     = (Empty file Name)

S4 : identifying the purpose of each field

S5: Map required fields to fact table

Load properties

S1: Set Data set properties S2: Set oracle properties Table name = dim 1 Write method = upset  (update and insert) Upset order = update then insert lab :-  SCD – 1 Oracle enterprise 0 load  dim 1 Oracle enterprise 1  Load              src Sno                    Sname 111                    shilpa 222                 repeka 333                  Archama SCD

1. Select output link as fact
2. Map the look up key sno
3. Source type = flat file

Source   name  = E :/shilpa/scd1.txt     empty file

1. Map required fields to fact  table

Compile and RUN (b) Data set ----------------------àoracle enterprise à In table  (a) , save the columns is Data set 1 and also copy the path of div 1 data set Now, is this job Data setpaste the path is the file  columns load  (SCD)   For indepth understanding of DataStage click on

• Introduction to DataStage
• Architecture of Data Stage
• Oracle Enterprise in Data Stage
• SCD(Slow changing Dimension) in Data Stage
• DataStage Tutorials