Skip to main content

Table 6 ADS-HC running time with different granularity of data segmentation

From: ADS-HCSpark: A scalable HaplotypeCaller leveraging adaptive data segmentation to accelerate variant calling on Spark

Execution time (min)

only sorted

2 blocks + sorted

4 blocks + sorted

6 blocks + sorted

8 blocks + sorted

0 blocks + no sorted

D1 (1 node)

73.89

75.30

75.76

76.42

77.47

85.09

D2 (1 node)

100.14

101.05

103.79

105.12

103.44

106.68

D3 (1 node)

71.45

71.43

72.67

73.30

75.11

80.67

D4 (1 node)

160.59

161.40

164.32

165.93

168.43

182.85

D1 (6 nodes)

27.32

20.34

19.46

18.41

18.49

33.92

D2 (6 nodes)

33.02

28.95

27.15

26.01

26.22

40.32

D3 (6 nodes)

26.21

20.96

19.31

19.33

20.65

33.14

D4 (6 nodes)

47.38

44.42

43.75

43.22

43.22

62.13