September 26, 2021

Part 1: Informatica MDM Interview Questions

What do you understand about MDM?
- MDM is an acronym for Master data management. 
- MDM is a methodology of allowing an organization to link all of its critical data to one file, which is called a master file. This file acts as a single point of reference to make important business decisions. 
- MDM acts as a central repository of data sharing between various departments when done properly.
- When implemented properly, MDM networks data sharing among individuals and enterprises.

What is Data Warehousing?
- Data Warehousing (DW) is a method of gathering and managing data from multiple sources to help organizations with valuable insights. 
- Data warehousing is the central source for the BI tools and for visualizing data.
- The Data warehouse transforms the data into understandable information and makes it available for business users.

What are the various stages of Data Warehousing?
- Offline Operational Databases: Perhaps this is the first stage in which a data warehouse system is developed from copying the operational process into an offline server. This process doesn’t make any impact or disturbance to the actual performance of the system.
- Offline Data Warehouse: In this stage, the operational data gets updated into the warehouse on a timely basis like daily, weekly or monthly. Also the data gets stored in an integrated report oriented way.
- Real-Time Data Warehouse: In this stage, data warehouses are updated whenever an event or transaction happens. A transaction or event includes an order or a booking or a delivery etc.
- Integrated Data Warehouse: In this stage, transactions and activity generated by warehouses go through the operating system and are helpful in the daily functioning of a business.

What is Dimensional Modelling?
Dimensional Modelling has two types of tables which are distinct from the third normal form. Here, Facts table containing the measurements of the business and Dimension table containing the context viz. dimension of calculation of the measurements are implemented.

What is Informatica PowerCenter?
PowerCenter is the data integration software of Informatica Corporation. The usage of this software allows loading the data into centralized locations such as data warehouses. Data taken from multiple sources can be used as per the business requirements.

What are the components of Informatica PowerCenter?
PowerCenter Domain
PowerCenter Repository
Administration Console
PowerCenter Client
Repository Service
Integration service
Web Services Hub
Data Analyzer
Metadata Manager
PowerCenter Repository Reports.

Define the terms: Mapping, Mapplet, and Transformation.
- Data mapping is a process of mapping a field data source to the targeted file or location. There are multiple data mapping tools available which help the developers in mapping the data from a source file to target file.
- A transformation is an object that generates, modifies or passes data. So, it is a repository object. Transformations in a mapping represent the operations the Integration Service performs certain operations on the data in a mapping that are represented. Transformation ports are linked in a mapping or mapplet and the data passes through these ports.
- A Mapplet is a reusable object containing a set of transformations. It enables to reuse that transformation logic in multiple mappings.

List the objects that are not allowed to be used in Mapplet?
- COBOL source definition
- Normalizer transformations
- Joiner transformations
- Pre or post-session stored procedures
- sequence generator transformations that are non-reusable
- XML source definitions
- Target definitions
- IBM MQ source definitions
- Power mart 3.5 styles of Lookup functions

What is Fact and Dimension Table?
- In data warehousing, a fact table contains metrics, measures or facts about a business process. 
- The fact table is located at the snowflake schema or star schema surrounded by multiple dimension tables. 
- A fact table typically contains two columns in which one contains facts and the other one is a foreign key.
- E.g in the 'milk production' business, the 'average milk produced by a single cow', or 'weekly milk production' would be considered as a measurement of the business process. Whereas textual attributes of measurements stored in the facts tables are included in the dimensional table.
-A dimensional table is a table in the star schema of a data warehouse. 
- While building Data Warehouses dimensional data models use dimension tables and facts. 
- The dimension table is a compilation of hierarchies, categories, and logic.

What are the foreign keys in dimension and facts tables?
- The primary keys of entity tables are called foreign keys of dimension tables.
- The primary keys of dimension tables are foreign keys of facts tables.

What is Data Mining?
- Data Mining is the process of analyzing data from different perspectives and summarizing it into useful information.
- It allows the users to find the previously unknown patterns and relationships between various elements in data. 
- The insights extracted for data mining would help in fraud detection, marketing, and scientific discovery, etc. 
- The other names for data mining are Knowledge extraction, Knowledge discovery, information harvesting, data/pattern analysis, etc.

What are the two ways of loading a Dimensional Table?
- Conventional (Slow) method:  Before the data gets loaded, all the keys and constraints are validated against it. Data integrity is maintained this way.
- Direct (Fast) way: Before the data gets loaded, all the constraints and keys are disabled. Data is validated against all the constraints and keys once it is loaded.

What are the ways to switch from one environment to another in Informatica?
- By copying objects/folders 
- The repository can also be exported and imported into the new environment. 
- We can also export each mapping to XML and import in a new environment. 
- Informatica deployment groups can also be used.

What is the difference between OLTP and OLAP.
- OLTP Stands for Online Transaction Processing, where as OLAP stands for Online Analytical Processing.
- OLTP is an application that modifies the data, the instance receives and it has a large number of recurrent users. OLAP is an application that collects, manage, process and presents the multidimensional data for management and analytics purposes

How can we erase a duplicate record in Informatica?
- In source, qualifier use selects distinctly.
- Override SQL query in Source qualifier.
- Use Aggregator and group by all fields.

What is the difference between mapping variable and mapping parameter?
- A mapping variable is dynamic in nature and changes through the sessions. The integration service saves the value of the Mapping variable in the repository on successful completion of every session. And the same value will be used when we run the session.
- A Mapping Parameter is a static value. We are required to define a variable before executing a session and the value we have given remains the same even after successful completion of the session. While executing the session Powercenter validates the value from the Parameter and keeps the same value till the end of the session. Whenever we run the session values are extracted from the file.

What are the various LOCK’s used in Informatica MDM 10.1?
- Exclusive Lock: This Lock can only allow access to a single user to make changes to underlying ORS and also blocks other users from modifying metadata in the ORS till the Exclusive lock exits.
- Write Lock: This lock allows multiple users at a time to make changes to the underlying metadata.

What is a Base object in MDM?
A base object is used to define core business entities such as products, employees, customers, accounts, etc.  
- It is an endpoint for consolidating data from various systems. 
- The Schema manager is the only way you have to define base objects, it is not allowed to configure in the database.

Explain various data loading stages in MDM?
Land, Stage, Load, Match and Consolidate

How to find all invalid mappings in a folder?
We can find the invalid mappings by executing below query:
SELECT MAPPING_NAME FROM REP_ALL_MAPPINGS WHERE SUBJECT_AREA= 'FOLDER_NAME' AND PARENT_MAPPING_IS_VALIED <>1

No comments:

Post a Comment