Infosource

Infosoures 
A non-persistent structure consisting of InfoObjects for joining two transformations.

You always use an InfoSource when you want to perform two (or more) transformations consecutively in the data flow - without additional storage of the data.

My Notes 

Having read articles on Infosource, I am not able to completely understand the use. That might be because I didnot work on bw 3.5 where Infosources are mandatory and with out them, we cannot connect flows. Having worked from BW 7.0, I dont find them useful, as every where they say, we can use it to connect different datasources to a DSO, I would use a direct transformation.

Useful Links:

http://help.sap.com/saphelp_nw70/helpdata/en/7e/001342743eda2ce10000000a1550b0/content.htm

Compounding

compounding attribute are nothing jst a combinaiton of two char to make a unique one i mean none of them is existant without the other or in other words there is no meaning if the are separated

Typically in a organization the employee id are allocated in serial like say 10001, 10002 and so on. Lets your Organization comes out with a new employee id scheme where the employee id for each location would start with 101. So the employee id starting for india would be india/101 and for US would be US/101. Now note that the employee india/101 and US/101 are different. Now if someone has to contact employee 101 he needs to know the location without which he cannot uniquely identify the employee. Hence in this case location is the compunding attribute.

A maximum of 13 characteristics can be compounded for an InfoObject. the characteristic values can also have a maximum of 60 characters. This includes the concatenated value, meaning the total length of the characteristic in compounding plus the length of the characteristic itself.

Data Loading

Lo related datasources.

• New Datasource
• Checking : Check the existing setup table to see if data is present in the setup tables. Go to SE16 and give the setup table name
• MC<Application Number>M_0<datasource>SETUP
• MC02M_0ITMSETUP
• Deleting : If data is present in the setup tables, delete them and re-load them again. Remember  that set up tables are filled for the whole application and not for each datasource.
• Set up table data will be deleted for whole 'Purchasing' if deleting for '02' even though you want data reload for only Purchasing Item datasource 2lis_02_itm.
• T-Code : LBWG 
• Or SBIW -> Settings for Application-Specific Datasources(PI) -> Logistics  -> Managing Extract Structures -> Initialization -> Delete the contents of the setup tables  
• Filling : Go to corresponding t-code for each application (unlike deleting, filling setup table t-code is different for each application or Goto SBIW ->  SBIW -> Settings for Application-Specific Datasources(PI) -> Logistics  -> Managing Extract Structures -> Initialization -> Filling in setup tables -> Application Specific setup of statistical data -> (choose your application)
• Depending on the screen you get over there, selections can be given and data can be filled to the setup tables or you can run it wide open.
• Remember to run in background, so that its easy to monitor and know the status. The job will be scheduled under job name 'RMCENEUA'.
• Once the job is completed, the data is available to load to BW Via 'Full Update'
• Existing datasource
• Once the new field is added to the datasource, there are two ways to go.
• Historic data is not required for the new field
• So we need not do anything. Just add the field and the code if needed any in the exit and  let the regular process chains get the delta and fill the data
• Historic data needed
• For this we need to do a full load. So we need to follow all the steps as for a new datasource and fill the history.
• Looks for downtime ( when the delta's are not running) Fill the setup tables and do a full load to BW and to the infoprovider. It will take care of the existing data and also writing data to the new fields.
• Delta need/will NOT be disturbed. No Initialization is required. 

BI Content

Steps to follow while installing BI content
-----------------------------------------------------
Select the objects which are needed and drag them to the right side.

In grouping , make sure you are doing only necessary objects.

If upwards and downwards data flow is needed  (transformations and dtps etc), identify the object from the list , go back and search for the object in the left side - drag it individually and install it.

Go for Tree View instead of Hierarchy View ( which is the default view ) , to see a better view of the objects.

Remember, clicking on the parent, will select all the objects below it, even though they are not selected. you need to take care and select individual objects to install them.

Right click and select 'Donot Install any below' to make sure.

Better option would be to get the technical names of the individual transformations or dtps, whaterver needed from the flow and then install each object individually searching for them in the contect. This way, you are installing only necessary objects.

Datasources

Replication :
Goto RSDS -> Datasource and Source System -> Replicate

Infopackages

These are used to fetch data from ECC to BW. There are different update modes which can be used when fetching data from the source system.

Above is the picture of the various tabs of an infopackage.

• The data selection tab is the place where we can add filters when extracting data from the source. The filters needed to be maintained in the source system to be seen here.
• The extraction tab gives the details of type of the datasource (*needs more appropriate explanation)
• Processing tab gives you the detail till what level the data will be processed, like till psa, data targets only.
• Data Targets gives the detail of all the objects for which data needs to be loaded.
• Update tab is the most important tab as it has details on how the datasource should work. More details are given below.
• Schedule is for triggering the data extraction

Update Tab
This is the important tab in the infopackage. here, we mention the type of update.

Full Update

When you run the using Full Upload, what ever data is there in source everything will be pulled to BW. If you use this, you can't run the Delta as there is no pointer set to find delta records. Only after initializing the datasource, delta records can be fetched.

A full update requests all data that corresponds to the selection criteria you determined in the scheduler. In a certain period of time specified in the scheduler, 100,000 data records are accumulated. With a full update, all 100 000 data records are requested.

When running the full loads, if we have more down time, we can fill setup tables and do a ful l load, else, we can do an Intialization without data transfer which doesn’t fetch any data but it places a pointer for delta and the next time the data is extracted, it fetches the full load records and delta records.

Initialize Delta Process

Only after doing the initialization with any of the below options, the datasource is made delta capable and can be seen in RSA7. The option of Delta is Infopackage is also available only after this process.

• Initialization with Data Transfer
• When executed, this will create an entry in the RSA7, which means the pointer is set for delta records. It also fetches all records from ECC and after its completed , it places the pointer. It is Full Load + Initialization.
• Initialization without Data Transfer
• When executed, this will create an entry in the RSA7, which means the pointer is set for delta records. It fetches a single record to BW which is a header record.
• Early Delta Initialization
• With early delta initialization, you have the option of writing the data into the delta queue or into the delta tables for the application during the initialization request in the source system. This means that you are able to execute the initialization of the delta process (the init request), without having to stop the updating of data in the source system.
• You can only execute an early delta initialization if the DataSource extractor called in the source system with this data request supports this.

Delta

A delta update only requests data which has appeared since the last delta. Before you can request a delta update, you must first initialize the delta process.  A delta update is only possible for loading from SAP source systems.  If a delta fails (status ’red’ in the monitor) or the overall    status of the delta request has been set to red manually, the next data request will be performed in repeat mode.

• F : Flat file provides the delta
• E : Extractor determines the delta, Ex: LIS, COPA
• D: Application determines the delta, Ex: LO, FI-AR/AP
• A: Use ALE change log delta

Full Repair  Request

This is present in the ‘Scheduler’ menu -> Repair Full Request, on the top when you open the infopackage.  These are used to fill in the gaps where delta extractions did not extract all delta records , when delta extraction process failed and cannot be extracted/restarted.

The request gets data from setup tables without delta being disturbed, wherein the setup tables are filled only according to the criteria.

These are also useful when you are initially extracting large volumes of data, whereby you execute an Init w/o data transfer and execute multiple Infopackage (Only Full Update) that are full repair request with specific selection criteria.

Finding the Infopackage Job

Once the infopackage is triggered for running, if you want to see the job details,

Environment -> Process Overview -> In the Source System

Else,

Copy the request number , go to the source system SM37 Add 'BI<request number>' and user as '*'

Data transfer using Infopackages

Data transfer using infopackes is done using IDocs and TRfc. Check LINK for more info
Infopackages can fail due to a number of reason.
Some of the common issues can be found : LINK