Skip to main content

Table 2 Classification speed of CLARK, Kraken, and cuCLARK at genus-level measured in terms of 106 classified sequences (reads) per minute

From: Accelerating metagenomic read classification on CUDA-enabled GPUs

 

CPU threads

GPUs

HiSeq Speed

MiSeq Speed

wgsim200 Speed

wgsim250 Speed

art454 Speed

fqIon Speed

CLARK

1

-

1.66

0.75

0.65

0.51

0.26

0.68

CLARK

16

-

13.46

8.96

8.07

6.52

3.9

9.4

Kraken

1

-

2.37

1.33

1.15

0.89

0.64

1.32

Kraken

16

-

12.6

13.74

14.45

11.95

8.75

15.48

cuCLARK

8

1

49.27

41.22

37.23

33.89

25.13

39.93

cuCLARK/CLARK

  

3.7

4.6

4.6

5.2

6.4

4.2

cuCLARK/Kraken

  

3.9

3

2.6

2.8

2.9

2.6

  1. The two bottom rows report the speedups of cuCLARK over CLARK and Kraken executed with 16 CPU threads