Skip to Content

SAP BI Introduction and Terminology

Roles and Responsibilities of SAP BI consultant:

  1. Modeling
  2. Extraction (ETL)
  3. Reporting
  4. Performance Tuning
  5. Broadcasting

Products of SAP

  1. SAP R/2
    1. Maintenance system
    2. Heavy load on Database
    3. Slow process
  2. SAP R/3
    1. Normal System
    2. Client–>Application Server–>Database Server
    3. Enterprise Central Component 6.0
  3. MySAP
  4. SAPMini
  5. IS-Industry Specific SAP tools like retail, oil and gas

SAP Netweaver Platform

  1. SOA: Service Oriented Architecture
  2. ESA: Enterprise Service Architecture

 

Any Application that provide end to end solution is called SOA or ESA.

SAP Netweaver Achitecture

SAP-IDoc is Document format of SAP

XI-Exchange Integration

Modeling:

  1. Developing target table
  2. InfoObjects:
    1. Fields and Naming Conventions
    2. Headers of table
  3. InfoCubes:
    1. Multidimensional table (slice and dice)
  4. DataStore Objects (DSO):
    1. 2-D tables
  5. MultiProviders:
    1. View
    2. Union of tables
  6. InfoSets:
    1. Joins of tables
    2. Intersection of tables

1-2 are header of BI

3-6 are InfoProvider of BI

Extraction:

  1. Development of link from source to target tables
  2. DataSource: Selected data and structure
  3. Transformation
  4. Persistent Staging Area (PSA): 2D intermediate table for error handeling
  5. Data Transfer process to target table

Reports (Business Explorer-BEx):

  1. Analyser: Excel format
  2. Query Designer
    1. According to requirement of end user
    2. input: InfoProvider
    3. output: web page
  3. Report Designer
    1. Formatted ouput
    2. input: Query
  4. Web Application Designer (WAD)
    1. assigning input to web page
      1. data binding
    2. input:
      1. Query from Query Designer
      2. InfoProvider
      3. Query view from Query Analyser

Performance Tuning

  1. Aggregates: Small Cubes
  2. Compression
  3. Partitions
  4. Index
  5. Line item dimension
  6. OLAP Cache
  7. BI Accelerator index

Broadcasting

  1. When to send: Time Scheduler
  2. How to send: Format
  3. Where to send: Location

Types of Application

OLTP (R/3) OLAP (BI)
Online transaction process Online Analytical Process
Updating data immediately Updating data less frequent
Relevance of data: Present Historical Data
Amount of data for transaction: Small Huge
Data in the DB: Normalized De-normalized
No of user: More Few User

Extraction Process involved in SAP BI

Update Mode:

Parameter or setting which has to be set before extracting data to BI which defines what amount of data has to be extracted from source

  1. Full update
  2. Initial update
  3. Delta Update: Modified and new data are updated

Database Design:

  • R/3 Database design
  1. 2D model of table
  2. Key columns and non-key columns
  3. Key field or non-key fields
  4. Composite key: Combination of more than one key become primary key
  5. Foreign key: primary key is used in other tables
  • ER model (Entity Relationship Model)
    • Any transaction take place, this data will not be updated in a single table. That transaction has been split (based on relationship between entities) and updated into multiple table to avoid redundancy.
    • R/3 is normalized, optimal for storage not for analysis.
    • R/3 is de-normalized, optimal for analysis not for storage.

BI Concepts

  • InfoObjects: Fields, Headers of tables/columns
  • Types of InfoObjects:
    • Key Figures:
      • The object which we analyse in the report has to be define under key figure
      • The key figures are the measurable objects
      • Key figure always numeric
      • All numeric objects are not key figure
    • Character:
      • On what bases we analyse the key figure are character
      • Primary key will always character
    • Unit Character:
      • Unit character will give value to the key figure
      • Unit character will always follow where key figure are used
    • Time Character:
      • It will called values related date/time of transaction.
      • Which include year, month, day and time
  • Types of Data:
    • Master Data:
      • The detailed level of information of character infoObjects is called as master data
      • This data is nothing to do with occurrence of business
      • Master data does not change frequently
    • Transactional Data:
      • Data related to occurrence of business is called transactional data
      • It will change frequently
  • Behaviour of data:
    • behaviour of Master data:
      • overwrite of data
      • It gives the presents truth
    • Behaviour of Transactional data:
      • Never overwrite of data
      • Addition of data
      • It gives the fact truth
  • Types of Master data
    • Attributes:
      • Whenever you represent master data in such a way that if all other InfoObjects are frequently depending on primary key InfoObjects
    • Texts:
      • It represents language and description
    • Hierarchy:
      • It represents master data level up and level down
      • Redundant of data
      • Extraction of data from R/3 to BI directly is not possible
  • Types of Transaction data
    • Actual: R/3
    • Planned: BI

Steps in modeling

  1. Identify no of InfoObjects to be created in BI
  2. Identify each InfoObjects as what type has to be created
  3. Identify what type of data are extracting (Master of Transaction)
  4. Identify what type of Master data you are extracting (Attribute, text or Hierarchy)
Be the first to leave a comment
You must be Logged on to comment or reply to a post.