SAP BI/BW LO - EXTRACTION

Definition : 

      Lo extractor is used to extract logistics data from sap r/3 to sap bw.

In older days LIS used in place of LO.

Differences of LIS and LO:

 Both are used to extract the logistics data.

LIS :

1. LIS uses transparent tables. 
2. LIS uses v1 and v2 update modes only.
3. LIS is customer generated extractors.
4. LIS generates transfer structure.
5. We need to create data sources and everything.
6. Two transparant tables used in delta mechanism. 

 LO :

1. LO uses cluster tables. 
2. LO uses v1 and v2 and v3 update modes. 
3. LO is bw content extractor. 
4. LO generates extract structure. 
5. In LO we have pre-defined data sources. 
6. Extraction queue,Update queue,Delta queue used  in delta mechanism.   

      Major difference is LO provides  data sources  specific to level of information for ex : 2LIS_02_HDR purchasing order header level data,2LIS_02_ITM purchasing order item level data,2LIS_02_SCL purchasing order schedule level data.But in case of LIS  it is specific to the particular application.

FUNCTIONS OF LO - COCKPIT (LBWE)
 1. Maintenance of extract structure.
 2.Maintenance of data sources.
 3.Activating updates.
 4.Controlling updates.

• Direct Delta.
• Unserialized V3 Update.
• Queued Delta.  

In LO - Cockpit we use setup tables to Init/Full upload.These are also called cluster tables.

UPDATE MODES : 

V1 -- Synchronous update..It will transfer the data in synchronous manner means if  user A and user B are posting the documents.For ex user A completes the document than before B completion..User A submitted the document... after it is successfully posted into specific queue  only the next document  will be posted into queue. For suppose the document submitted by the user A fails then the documents from the other sources will wait untill it gets resolved. 

V2 -- Asynchronous update..Even if the document posted by the user A fails the documents from the next users will be updated to the specific queue..no matter which document submitted first.

V3 -- Asynchronous batch v3 update ..The documents will be posted in background in the specific time period. V3 is specially for bw extraction.

Direct Delta :

 When we do info package with full load the data will be taken from setup table.So we need to fill the setup table (It also called statistical setup) by using the t-code OLI3BW for purchasing order data.Before going to do the statistical setup  we need to delete the data from the setup table using t-codes LBWG  or SE14 .

LBWG -- Used to delete the data from entire application for example 02(purchasing),03(inventory),11(sales) etc....

SE14 -- Used to delete the data from specific data source for example 2LIS_02_HDR (from purchasing document header level data) , 2LIS_11_VAITM (from sales document item level data) etc...

Q ) Why do we need to delete the data from the setup table?

A) Because it may contain some garbage value.

First we need to lock the transactions using t-code SM01 .

 ME21(create),ME22(change),ME23(display).

Accessing the application tables are not permitted  hence setup tables are there to collect the required data from the application tables.

 Enter t-code OLI3BW for filling the setup table with purchasing document data(02).

Setup table gets the data from the appln.tables(vbak,vbap,ekko,ekpo etc..) through comminication structure as shown in the above figure.Then the entire data will be loaded to the datasource from setup table.. through extract structure when u do infopackage with full load. 

In Direct Delta when u do the document posting the data records(new or modified) directly posted into BW Delta Queue(RSA7) and  into the Application tables at the same time.

Advantages:

   1. Using this we can send max. upto 10000 data records.

   2. No need to schedule the V3 job control to transfer the data into bw .

Disadvantages:

     1.For each document posting the data will be stored  in LUW of bw delta queue.

     2.Different document changes are not summarized into single LUW ,so number of LUW's are significantly  increased for each document posting.So it decreases the performance of the OLTP system. 

    So this method only  recommended for the customers with  low occurances of documents.  

Unserialized V3 Update : 

       

 When the document is posting the data will be loaded into application tables(v1) and update queue instead of  storing into bw delta queue (v2).

  Using v3 job control the data will be posted to bw delta queue(in specific time period).

Advantages :

  

    1.We can load upto 100000 records.

    2.Using v3 job control we can schedule the time period for data transfer.
    3.This is recommended when we want to send the data to Info cubes or Dso's with additive mode.

Disadvantages :

     Serialization is not ensured.

Queued Delta :

    

       

 When document is posting the data records will be posted into extraction queue and update in bw delta queue using v3 job control.

Advantages :

      1.Supports multiple languages.

      2.We can load more than 100000 records.

      3.Serialization is possible.

      4.Changed or Delta records are summarized into single LUW so it increases the OLTP performance.

      5.Using v3 job control we can schedule the time period for transfer the data to bw delta queue.

Q) What is LUW?
    It is a logical unit of work.It works like a changelog table in the dso and it maintains new image N,before image X,after image ' '  for changed/modified records.