Assignment No. 4 Graded Semester Fall 2015 Data Warehousing– CS614 due date 8 Feb,2016

[+•°¯`•• βĮЎÃ (ÃβẸẸŘÃ) ••´¯°•+]

Objective: The assignment has been designed to develop your ability to calculate the Bitmap index.  Instructions: Please read the following instructions carefully before solving & submitting assignment:   1.      The assignment will not be accepted after due date in any case (whether it is the case of load shedding or emergency electric failure or internet malfunctioning etc.). 2.      Zero marks will be awarded to the assignment that does not open or the file is corrupt. 3.      The assignment file must be an MS Word (.doc) file format; Assignment will not be accepted in any other format. 4.      Zero marks will be awarded to the assignment if copied (from other student or copied from handouts or internet). 5.      Zero marks will be awarded to the assignment if the Student ID is not mentioned in the assignment file.   For any query about the assignment, contact only at [e-mail] Do not post queries related to assignment on MDB.   GOOD LUCK Question 1                                                                                                                                         [10 Marks]   Consider the following table:   Player_Team Player_ID Player_Name Team_ID Pool_ID IND-06 Gangoli IND A AFG-05 Najeeb AFG B SA-07 AB Devillier SA A AU-01 Steve Waugh AU B IND-01 Tandulker IND A AU-04 Maxwell AU B AFG-01 Nawroze AFG B SA-09 Dal Styn SA A   Consider the following query:   SELECT Count (Player_Team.Pool_ID) AS CountOfPool_ID FROM Player_Team GROUP BY Player_Team.Pool_ID;   Answer the following questions:   You need to identify the number of clusters from this data. Secondly, you have to identify whether the given clustering is one way or two way clustering. Your answer should support by valid reasons.   Question 2                                                                                                                                         [10 Marks] Consider the following tables:   Player   Player_ID Player_Name Team PK-01 Wasim Pakistan PK-02 Misbah Pakistan SA-03 AB Devillier South Africa   Award   Award_ID Match_ID Player_ID 01 01 PK-01 01 02 PK-01 02 03 PK-02 01 04 SA-03   Consider the following query:   Select * from Player P, Award A where P.Team= ‘Pakistan’ and A.Award_ID = ‘01’ and P.Player_ID = A.Player_ID   Suppose this query is executed using Naive Nested-Loop join and (i.e. there is no index created on both Player and Award tables). Mention that which table should be the Outer table to get minimum I/O by manually calculating the cost in both cases i.e. when “Player” is outer table and when “Award” is outer table.     Note: You need to mention the calculations in your solutions where required.

[Naeem MCS (4th)]

Player_ID	Player_Name	Team_ID	Pool_ID
IND-06	Gangoli	IND	A
IND-01	Tandulker	IND	A
SA-07	AB Devillier	SA	A
SA-09	Dal Styn	SA	A
AU-01	Steve Waugh	AU	B
AU-04	Maxwell	AU	B
AFG-01	Nawroze	AFG	B
AFG-05	Najeeb	AFG	B

[Naeem MCS (4th)]

this is the answer of question no.1

[Jamil Asif+ GUA]

Complete Solution....

Is it right?

[Qamer]

identify number of clusster ?

[Aabish Hameed]

guys plz give solution of 2nd question. 

[▌general♥]

Idea Solution................. 

[+ M.Tariq Malik]

Question no. 1:

Consider the following table:

Player_Team

Player_ID	Player_Name	Team_ID	Pool_ID
WI-06	Richordson	WI	A
PK-05	Misbah	PK	A
SA-07	AB Devillier	SA	B
AU-01	Steve Waugh	AU	B
PK-01	Hafiz	PK	A
AU-04	Maxwell	AU	B
WI-01	Ambrose	WI	A
SA-09	Dal Styn	SA	B

 

Consider the following query:

SELECT Count (Player_Team.Pool_ID) AS CountOfPool_ID

FROM Player_Team

GROUP BY Player_Team.Pool_ID;

Answer the following questions:

1. You need to identify the number of clusters from this data.
2. Secondly, you have to identify whether the given clustering is one way or two way clustering. Your answer should support by valid reasons.

 

 

 

 

 

 

 

Answer:

Cluster indexing on Team_ID

Player_ID	Player_Name	Team_ID	Pool_ID
WI-06	Richordson	WI	A
WI-01	Ambrose	WI	A
PK-05	Misbah	PK	A
PK-01	Hafiz	PK	A
SA-07	AB Devillier	SA	B
SA-09	Dal Styn	SA	B
AU-01	Steve Waugh	AU	B
AU-04	Maxwell	AU	B

 

Cluster indexing on Pool_ID

Player_ID	Player_Name	Team_ID	Pool_ID
WI-06	Richordson	WI	A
PK-05	Misbah	PK	A
PK-01	Hafiz	PK	A
WI-01	Ambrose	WI	A
SA-07	AB Devillier	SA	B
AU-01	Steve Waugh	AU	B
AU-04	Maxwell	AU	B
SA-09	Dal Styn	SA	B

 

In this Query the indicate the separate grouping of objects because all the object separately. It’s a one way cluster.

 

Player

 

Player_ID	Player_Name	Team
PK-01	Wasim	Pakistan
PK-02	Misbah	Pakistan
SA-03	AB Devillier	South Africa

 

Award

Award_ID	Match_ID	Player_ID
01	01	PK-01
01	02	PK-01
02	03	PK-02
01	04	SA-03

 

Consider the following query:

 

Select * from Player P, Award A where P.Team= ‘Pakistan’ and A.Award_ID = ‘01’ and P.Player_ID = A.Player_ID

 

Suppose this query is executed using Naive Nested-Loop join (i.e. there is no index created on both Player and Award tables). Mention that which table should be the Outer table to get minimum I/O by manually calculating the cost in both cases i.e. when “Player” is outer table and when “Award” is outer table.

 

 

Answer:

 

Qualifying blocks in table P = 3

Size of Table P=9 Blocks

Qualifying blocks in table A = 4

Size of Table A=12 Blocks

 

If table_P is outer & table_A is inner

Formula for Join cost = size of blocks of table_P + (table blocks of table_P * size of blocks of table_A

 

Join cost1(player_table:outer, Award_table::inner)

=9+(3 * 12) =9+36

=45

{ size of table_P +(blocks of table_P * size of table_A)

 

If table_A is outer & table_P is inner

Formula for Join cost = size of blocks of table_A + (table blocks of table_A * size of blocks of table_P

 

Join cost2(Award_table:outer, Player_table::inner)

=12+(4 * 9) =12+36

=48

{ size of table_A +(blocks of table_A * size of table_P)

 

[Asif Raza]

Thanx to all of u guys...

U r really helpful.

[2wn3cpyldgot0]

All solutions are wrong............. 

[▌general♥]

Read comments on the files mentioned above carefully, these are idea solutions.........

[2wn3cpyldgot0]

bro menay apka nai kaha i know ap nay just idea dia sab nay asay hi copy paste krlena so kaha k sab wrong samj k kerain and discuss kren hojayegi complete

[Qamer]

tu thek dy do