Skip to main content

Table 2 Comparison of runtime (in seconds) between multi-threaded MIBE (4 threads) and CUDA-MI

From: Parallel mutual information estimation for inferring gene regulatory networks on GPUs

ID

MIBE

CUDA-MI

Speedup

nne2048_911

220.87

14.78

14.9

nne4096_911

867.35

45.17

19.2

nasc2048_2996

2174.65

44.89

48.4

nasc4096_2996

8645.28

153.46

56.3

Yeast_6221_80

246.01

30

8.2

Yeast_6307_215

490.46

42.18

11.6

S2000_1000

229.52

14.58

15.7

S4000_1000

904.86

50.89

17.8

S8000_1000

3635.11

192.05

18.9

S4000_3000

5960.87

148.09

40.3

S4000_4000

7993.99

199.07

40.2

S4000 5000

13790.19

256.76

53.7

S10000 _2000

34315.69

520.12

66.0

S10000 _3000

51705.18

781.73

66.1

S10000 _4000

68799.74

838.74

82.0