Skip to main content

Table 11 SSB execution times (in seconds): partitioning by “Od_Year” and “S_Region” and bucketing by “Suppkey”

From: Evaluating partitioning and bucketing strategies for Hive-based Big Data Warehousing systems

 

SF = 30

SF = 100

SF = 300

SF = 30

SF = 100

SF = 300

HIVE

PRESTO

SS

SS-PB

SS

SS-PB

SS

SS-PB

SS

SS-PB

SS

SS-PB

SS

SS-PB

Q1.1

25

19

31

22

44

27

5

3

13

5

36

12

Q1.2

24

25

29

32

42

51

5

8

13

21

34

65

Q1.3

24

19

29

22

43

25

4

2

13

5

35

12

Q2.1

32

28

47

43

531

50

8

5

19

12

59

37

Q2.2

31

26

46

41

531

160

7

5

18

10

51

27

Q2.3

30

26

44

41

531

45

7

4

17

9

49

26

Q3.1

35

25

59

36

651

66

8

5

29

14

81

44

Q3.2

30

30

45

50

677

92

6

9

17

31

51

100

Q3.3

33

36

219

78

665

78

5

9

15

25

43

86

Q3.4

34

33

222

226

675

81

6

10

15

30

43

91

Q4.1

38

33

86

70

226

127

13

9

43

24

119

65

Q4.2

49

30

70

57

141

60

9

4

26

13

69

25

Q4.3

34

31

54

47

116

72

8

7

23

21

63

60

Total

420

362

982

765

4874

933

92

81

262

220

733

650

DIF

 

− 14%

 

− 22%

 

− 81%

 

− 12%

 

− 16%

 

− 11%

  1. Italic values indicate the fastest processing time by query, workload, tool and data model