Teradata
It is an RDBMS
ROW 1 ROW 2 ROW 3 ROW 4 ROW 5 ROW 6
AMP 1 AMP 2 AMP 3
VDisk VDisk VDisk
ROW 3 ROW 2
ROW 1
ROW 4 ROW 5
ROW 6
Suppose we run the Query
SELECT *
FROM EMPLOYEE
WHERE ID = 100
#Parsing Engine
- It compiles our query
- Check “From EMPLOYEE”, Table is existed or not in Database
- Syntactically Query is right or wrong
In short,
- Validation
- Existence
- Execution Plan
All responsibility of “Parsing Engine”
#Message Passing Layer (Also Called Bynet)
- It is like ‘Highway’. It carries the execution plan
- It is like mediator
#AMP (Access Module Processor)
- Every AMP has its own Virtual Disk (VDisk) associated with it
- AMP as job do,
Data Storing and Data Retrieving
As see in figure we have 6 Row it evenly distributed data in AMP
Suppose we Run an query which is responsible for retrieving ROW3 then Parsing Engine make an execution plan to it AMP2 (as in fig.)
Message Parsing Layer (Bynet) carry the execution Plan
AMP2 send data via Bynet
Why Teradata is famous?
We can improve the performance very drastically just by changing the no. of AMP’s (For more AMP we need to pay for it)
Also, we can do scripting in Teradata
(i) FLOAD
(ii) MLOAD
(iii) TPUMP
In informatica to load a file in Database we do Mapping
In Teradata we can do same using Script in FLOAD
By Default, Teradata create an Index
CREATE MULTISET/SET TABLE EMPLOYEE
(
MULTISET: It allows Row Level Duplicate
SET: It doesn’t allow Row Level Duplicate
);
This is Sample Query of Create Table in Teradata.