• Exercise 1.1+1.2

– Linearization

–Why is order important?

6. Optimization

Data Warehousing & OLAP – Wolf-Tilo Balke – Institut für Informationssysteme – TU Braunschweig 1

11

36 47

22 53

78 89

64

911

1116 1217

1012 1313

1518 1619

1414

175

1910

2515

2720

Jan (1)

Feb(2)

Feb(3)

Feb(4)

D1

D2

• Exercise 2. a

6. Optimization

R1

R3

R5

08 Qtr3

08 Qtr4

09 Qtr1

Time

R

R1

R3

R508 Qtr4

09 Qtr1

Time

R

Data Warehousing & OLAP – Wolf-Tilo Balke – Institut für Informationssysteme – TU Braunschweig 2

R4

R6R2

08 Qtr1

08 Qtr2

08 Qtr3

Locationa b c d e f g

R6R2

08 Qtr1

08 Qtr2

08 Qtr3

a b c d e f g

R41 R42

• Exercise 2. a

6. Optimization

R1

R3

R508 Qtr4

09 Qtr1

Time

R

R1

R3

R508 Qtr4

09 Qtr1

Time

R

Data Warehousing & OLAP – Wolf-Tilo Balke – Institut für Informationssysteme – TU Braunschweig 3

R6R2

08 Qtr1

08 Qtr2

08 Qtr3

a b c d e f g

R6R2

08 Qtr1

08 Qtr2

08 Qtr3

a b c d e f g

R41 R42R41 R42

• Exercise 2. a

6. Optimization

R1

R31

R508 Qtr4

09 Qtr1

Time

R

R1

R3

R508 Qtr4

09 Qtr1

Time

R

R32

Data Warehousing & OLAP – Wolf-Tilo Balke – Institut für Informationssysteme – TU Braunschweig 4

R6R2

08 Qtr1

08 Qtr2

08 Qtr3

a b c d e f g

R6R2

08 Qtr1

08 Qtr2

08 Qtr3

a b c d e f g

R41 R42 R41 R42

• Exercise 2. a

6. Optimization

R1 R508 Qtr4

09 Qtr1

Time

R

DCR31

R5

08 Qtr3

08 Qtr4

09 Qtr1

Time

R

R32 R12

R11

Data Warehousing & OLAP – Wolf-Tilo Balke – Institut für Informationssysteme – TU Braunschweig 5

R6R2

08 Qtr1

08 Qtr2

08 Qtr3

a b c d e f g

BA

On X, highest minimum rectangles are B and D = ‘b’, and lowest maximum are A and C = ‘a’

On Y, highest minimum rectangle is D = ’09Qtr1’, and lowest maximum is A = ’08Qtr2’

Dx = 1/3; Dy = 3/5; => D and A will create the new split nodes

R6R2

08 Qtr1

08 Qtr2

08 Qtr3

a b c d e f gR41

R42

• Exercise 2. b

6. Optimization

R11 R2

R

R12

Data Warehousing & OLAP – Wolf-Tilo Balke – Institut für Informationssysteme – TU Braunschweig 6

R31 R41

R31.1 R31.2 R41.1 R41.2 R5.3 R6.3R6.2

R32 R42

R32.1 R32.2 R42.1 R42.2 R42.3

R5 R6

R5.2R5.1 R6.1

• Exercise 2. c

6. Optimization

R31

08 Qtr4

09 Qtr1

Time

R

R32 R12

R11

Data Warehousing & OLAP – Wolf-Tilo Balke – Institut für Informationssysteme – TU Braunschweig 7

R5

R6R2

08 Qtr1

08 Qtr2

08 Qtr3

08 Qtr4

a b c d e f gR41

R42

• Exercise 2. c

6. Optimization

R11 R2

R

R12

Data Warehousing & OLAP – Wolf-Tilo Balke – Institut für Informationssysteme – TU Braunschweig 8

R31 R41

R31.1 R31.2 R41.1 R41.2 R5.3 R6.3R6.2

R32 R42

R32.1 R32.2 R42.1 R42.2 R42.3

R5 R6

R5.2R5.1 R6.1

• Exercise 3.1

– UB-Trees

6. Optimization

Data Warehousing & OLAP – Wolf-Tilo Balke – Institut für Informationssysteme – TU Braunschweig 9

Data part

8 178 17 39 5139 51

2828

• Exercise 3.1

–Multidimensional Hierarchical Clustering (MHC)

6. Optimization

Sector Brown Goods White Goods

ALL

0 1Sector Brown Goods White GoodsBrown Goods White Goods

ALL

0 1

Data Warehousing & OLAP – Wolf-Tilo Balke – Institut für Informationssysteme – TU Braunschweig 10

...

Item

ProductGroup

Category VideoAudio

Camcorder VCR

TR-780 TRV-30 GR-AX 200 GV-500 SLV-E800

...

...

ID 2 11 5 8 21

00 01

0 1

Surrogate 00100000

0000 0001 0000 0001 0010

00100001 00110000 00110001 00110010

32 33 48 49 50

...

Item

ProductGroup

Category VideoAudio VideoAudio

Camcorder VCR

TR-780 TRV-30TR-780 TRV-30 GR-AX 200 GV-500 SLV-E800

...

...

ID 2 11 5 8 21

00 01

0 1

Surrogate 00100000

0000 0001 0000 0001 0010

00100001 00110000 00110001 00110010

32 33 48 49 50

• Exercise 3.2

– Multi-component BI:

• x = 4 *y+z, where 0 ≤ y ≤2, and 0 ≤z ≤3, called <3,4>

basis encoding

6. Optimization

Month Dec Nov Oct Sep Aug Jul Jun Mai Apr Mar Feb Jan

x = 4 *y+z, where 0 ≤ y ≤2, and 0 ≤z ≤3, called <3,4>

basis encoding

• 5 = 4*1+1

Data Warehousing & OLAP – Wolf-Tilo Balke – Institut für Informationssysteme – TU Braunschweig 11

Month Dec Nov Oct Sep Aug Jul Jun Mai Apr Mar Feb Jan

M A11 A10 A9 A8 A7 A6 A5 A4 A3 A2 A1 A0

5 0 0 0 0 0 0 1 0 0 0 0 0

X Y Z

M A2,1 A1,1 A0,1 A3,0 A2,0 A1,0 A0,0

5 0 1 0 0 0 1 0

• Exercise 3.2

– Range encoded BI

6. Optimization

Month Dec Nov Oct Sep Aug Jul Jun Mai Apr Mar Feb Jan

M A11 A10 A9 A8 A7 A6 A5 A4 A3 A2 A1 A0

0 1 1 1 1 1 1 1 1 1 1 1 1

– Persons born in March: ((NOT A1) AND A2)

Data Warehousing & OLAP – Wolf-Tilo Balke – Institut für Informationssysteme – TU Braunschweig 12

0 1 1 1 1 1 1 1 1 1 1 1 1

3 1 1 1 1 1 1 1 1 1 0 0 0

5 1 1 1 1 1 1 1 0 0 0 0 0

11 1 0 0 0 0 0 0 0 0 0 0 0