ODI Interview Questions and Answers
Q1. What is Oracle Data Integrator (What is ODI)?
Ans: Oracle Data Integrator (ODI), Oracle’s ETL tool, is a comprehensive records integration platform, capable of high overall performance, excessive extent data hundreds and statistics variations. It may be used to extract records from a diffusion of statistics assets (relational databases, flat documents, and so on.), perform complicated information modifications, and load statistics into a spread of target records stores (information warehouses, utility databases, flat files, and so forth.). ODI may be used to guide a selection of information integration initiatives, such as the subsequent:
Business Intelligence and Data Warehousing
Data Migrations and Data Consolidations
Master Data Management
Q2. What is the distinction between ETL and ELT?
Ans:In ETL we must have Middle Tier Server Engine wherein as in ELT shouldn't require Middle-Tier Server Engine. So it lessen cost. Network traffic is extra for ETL in which as for ELT much less community visitors examine to ETL
Q3. What is Repository and Types of Repositories?
Ans:A repository is a metadata matianance database person or schema. In this schema we will keep all odi additives facts like topology, security, designer and operator additives statistics.
There are sorts of repositories.
Master Repository
Work Repository
Q4. What is meant by means of OLAP and OLTP?
Ans:OLAP - OnLine Analytic Process - Maintain Historical Data
OLTP - OnLine Transaction Process - Daily Data (Business Data)
Q5. What is large distinction among Physical Schema and Logical Schema?
Ans:Physical Schema = Logical Schema + Context
For Example: Consider an Organization A whose Branches are A1,A2 and A3.
Consider the Structure of A1,A2 and A3 Schema's are same however placed in distinctive Servers. By the EOD all of the facts saved in A1,A2 and A3 to be saved in A.
For above scenario, developer develops one mapping with one logical Schema , 3 Physical Schema (A1,A2 and A3)and three Context (A1_CTX,A2_CTX and A3_CTX) . While executing the mapping if he selects A1_CTX it loads to A1. (i.E., Logical_schema+A1_CTX = A1... ) . That method we are able to reuse equal code to pick data from exceptional schema's.
Q6.What is an Agent and distinct sorts of Agents?
Ans:ODI agent is run time component which orchestrates the facts integration technique.
They are 3 sorts of Agents:
Standalone Agent - Light Weight
J2EE Agent - High Availability , Scalability, Security and higher manageability
Colocated Standalone Agent - Combination both standalone and J2EE agent
HubSpot Video
Q7.What is Context and it's cause?
Ans:Contexts bring together additives of the physical architecture (the real Architecture) of the facts device with components of the Oracle Data Integrator logical architecture (the Architecture on which the user works).
For example, contexts can also correspond to one of a kind execution environments (Development, Test and Production) or distinctive execution locations (Boston Site, New-York Site, and so forth.) wherein comparable physical aid exist.
Q8.What is Logical Schema?
Ans:Logical schema is alias to your bodily schemas and it's going to connect with your physical schemas throgh contexts. In Designer developer can get entry to most effective logical schemas he con now not access bodily schemas, developer can access logical schemas and context combination he
can connect bodily schemas.
Q9. What is Designer?
Ans:Designer issue we're the use of for whole development location. In design we will paintings for improvement like developing interfaces, packages, techniques,
variables, sequences and other gadgets, after finishing development we will generate schenarios and we will flow this all situations in to manufacturing.
Q10.What is Scenario?
Ans:A situation is a compiled or executable object and it is not editable, in odi manufacturing environemnt we are able to use best eventualities for mataining
security and hiding. Due to the fact this scenarios isn't always readeable and not editable. This scenarios simplest schedules or executable.
Q11. What is Interface ?
Ans:An interface we are the usage of to load facts from supply to target , right here we are able to practice all the differences and validations and aggregations.In odi 11g interface we will use most effective one goal in a single interface, if we want load information into a couple of goal tables we must go for multiple
interfaces and use IKM Multiple desk insert know-how module.
In ODI 12c Interfaces we're calling Mappings and we are able to use multiple target tables in a single mapping.
Q12. What is Yellow Interface?
Ans:Yellow interface we're the usage of for with out goal table and we will create dynamically goal table if we dont have target desk.For this we want to allow CREATE_TARGET_TABLE option=> TRUE at IKM stage.This yellow interface we can use for source table as well. This yellow interfaces we're calling Reusable mappings in 12c.
Q13. What is Simulation choice at interface?
Ans:This alternative we are using for to expose complete interface technique like LKM and IKM and CKM method like creating all $tables and placing into goal table queries everything it will supply a document.
Note: Simulation will supply simplest report it wont run in consultation or software.
Q14. What is CDC?
Ans:CDC alternate data seize we're the usage of for to extract changed facts from source. If source accepting Triggers on supply table we will use this CDC function if source machine they may be no longer allowing to create a triggers we can not use this CDC characteristic we will move for timestamp based filtering.
Note: If source desk having any timestamp or changed information column we can filtering if supply table dont have timestamp or changed date column we have to
use CDC feature (journalization idea) if they may be not allowing to create triggers on source desk we should choose everying and we need to preserve seperate
staging table for checking facts from supply and our staging table facts.
Note: If source table dont have timestamp column and they may be now not allowing to create CDc (triggers) then we want to create one staging desk STG_ORDERS
with timestamp column first we can load full statistics from supply desk (SOURCE_ORDERS) table to our stagning desk, but we can maintain this
staging data completely and we are able to create one more interface to load this STG_ORDERS desk to TARGET_ORDERS table based on timestamp column.
From next time onwards at the same time as choose facts from SOURCE_ORDERS desk we can check STG_ORDERS table if STG_ORDERS desk dont have source statistics
that records only we are able to pick out and we are able to insert.
SELECT * FROM SOURCE_ORDERS WHERE NOT EXITS ( SELECT 'X' FROM STG_ORDERS);
for this requirement we want two inerfaces one for SOURCE_ORDERS desk STG_ORDERS table and any other interface for
STG_ORDERS desk to TARGET_ORDERS table.
STG_ORDERS to TARGET_ORDERS table we can use timestamp filter situation on STG_ORDERS and it'll get selected only modified data.
We cannot delete information from STG_ORDERS desk.
Q15. What is SCD?
Ans:SCD Slowly converting size we are the use of for to load converting size information records.
In which are three forms of SCD'same
1) SCD TYPE1: No records, we will all columns SCD assets as OVERWRITE ON CHANGE in ODI.
SCD TYPE1: we will use IKM Oracle Incremental update KM or we can IKM Slowly Changing Dimension both we can use.
2) SCD TYPE2: Here we are able to hold no of times records, so we want to keep below SCD properties in SCD TYPE 2 desk.
1) ADD Row On CHANGE: this property will upload a row whenever we're receiving modified records to preserve history.
2) STARTING_ TIME STAMP: This will display document starting date and time
three) Ending Time stamp: this can display record finishing date and time
4) RECORD FLAG: this will display active record if it's miles Y and if it's far N for antique rows or history rows.
3) SCD TYPE three: As of now ODI isn't supporting we should customize our IKM Slowly Changing Dimension. It will preserve best one time records in different column like , CURRENT SALARY and PREVIOUS SALARY two columns.
Q16. What is Load Plan and extraordinary forms of Load Plans?
Ans:A Load Plan is an executable object in Oracle Data Integrator that could incorporate a hierarchy of steps that may be accomplished conditionally, in parallel or in series
Types of Load Plans
Parallel
Serial
Conditional
Q17. What is distinction between Package and Load Plan?
Ans:
Load Plans Packages
A Load Plan is an executable object in Oracle Data Integrator that may incorporate a hierarchy of steps that may be achieved conditionally, in parallel or in series There isn't any native aid
Load plans are moved to production as it is Packages want to transform into situation and then sell to production
We can create load plans in Execution Work repository additionally (like PROD Environment) We can’t create packages in Execution Work Repository (like production surroundings)
When we execute/restart Load Plans it's going to create new instances When we execute package it will create new session and for restart it's going to override the execute session
