recategorized by
3,478 views
2 votes
2 votes

The dependency preservation decomposition is a property to decompose database schema D, in which each functional dependency $X \rightarrow Y$ specified in $F$,

  1. Appeared directly in one of the relation schemas $Ri$ in the decomposed D.
  2. Could be inferred from dependencies that appear in some $Ri$.
  3. Both (A) and (B)
  4. None of these
recategorized by

2 Answers

3 votes
3 votes

Explanation:- The question itself requires a bit of explanation. It is not enough if you just know what is the right answer but you must also know why it is the right answer. The explanation would be a bit lengthy. Let us first dissect the question and explain some terms in terms of DBMS.

Decomposition - This means replacing a relation with a collection of smaller relations. 

Relation - Relation is known as Table.

Relation Schema - This is known as Table definition. Relation Schema for a "student" relation can be shown in the following way: 
Student(FirstName,LastName,DOB,Gender,Course,Regno,Address)

Definition of Dependency preservation decomposition:-
Each FD specified in F either appears directly in one of the relations in the decomposition, or be inferred from FDs that appear in some relation.

Let us consider an example for Dependency preservation

Let R be a relation R(A B C D) 
Let there be 3 functional dependencies. 
FD1: A->B
FD2: B->C
FD3: C->D
Let the relation R be decomposed into two more relations.
R1(A B C)  :  R2(C D)
Let us first consider the relation R1(A B C). Here between A and B the functional dependency FD1 is preserved. Between B and C, FD2 is preserved. 
Let us now consider the second relation R2(C D). Between C and D the FD, FD3 is preserved. So in the two relations R1 and R2, all the 3 functional dependencies are preserved.

Let us consider an example for Non-dependency preservation

Let R be a relation R(A B C D) Let there be again 3 functional dependencies.
FD1:A->B
FD2:B->C
FD3:C->D
Let the relation be decomposed into two more relations>
R1(A C D) R2(B C)
Let us first consider the relation R1(A C D). There is no FD between A and C. There is a FD3 between C and D.
Now let us consider the second relation R2(B C). There is FD2 between B and C.
So, the two relations only support only FD's FD2 and FD3. FD1 is not supported. So these relations does not preserve dependency.
Generally there are three desirable properties of a decomposition.

  1. Lossless
  2. Dependency preservation
  3. Minimal redundancy

The above question was based on dependency preservation decomposition. This example has been taken from the dependency preservation presentation by Jason Allen. The explanation is quite good there. 

SUMMARY:-

The dependency preservation decomposition is a property to be considered for decomposing a relation into two or more smaller relations. The functional dependency X->Y specified in F can appear directly in one of the relation schemas Ri in the decomposed D or it could be inferred from dependencies that appear in some Ri. So the answer for this question is C. 

Ans:-C

0 votes
0 votes

Answer: C

Both option A & B lists the proprerty of dependency preservation.

http://www.cs.sfu.ca/CourseCentral/354/zaiane/material/notes/Chapter7/node8.html

Related questions

0 votes
0 votes
2 answers
1
makhdoom ghaya asked Sep 7, 2016
4,260 views
Which of the following is an optimistic concurrency control method ?Validation basedTime stamp orderingLock-basedNone of these
0 votes
0 votes
3 answers
2
makhdoom ghaya asked Sep 7, 2016
2,796 views
In generalisation, the differences between members of an entity isMaximizedMinimizedBoth (A) & (B)None of these
3 votes
3 votes
1 answer
4
makhdoom ghaya asked Sep 11, 2016
2,094 views
One of the distinguished features of super-computer over other category of computer isParallel processingHighest accuracy levelMore speedMore capacity