| We have benchmarked a single CLC Genomics Machine against the most sold FPGA biocomputing solution. | |||||||||
| Algorithm | Technology used by CLC |
CLC Genomics Machine | Popular FPGA solution* | ||||||
| Smith-Waterman | SIMD | 70.1 GCUPS | 10.0 GCUPS | ||||||
| BLASTn | Standard | 6,800.0 GCUPS | 13,000.0 GCUPS | ||||||
| BLASTp | Standard | 104.0 GCUPS | 60.0 GCUPS | ||||||
| HMMsearch | SIMD | 7.8 GCUPS | 3.4 GCUPS | ||||||
| HMMpfam | SIMD | 8.1 GCUPS | 3.4 GCUPS | ||||||
SIMD: Single Instruction Multiple Data |
|||||||||
Performance |
|||||||||
7 |
Human genome de novo assembly 7 hours, 1 computer, 42 GB RAM, 3600 million reads |
||||||||
63 |
Human genome read mapping, ungapped 63 minutes, 83 million Illumina read, 36basepairs, 1 fold coverage |
||||||||
2.5 |
Human genome read mapping, gapped 2.5 hours, 83 million Illumina reads, 36 basepairs, 1 fold coverage |
||||||||
72 |
Human genome read mapping, gapped 72 minutes, 12 million 454 reads, 250 basepairs, 1 fold coverage, Dr. Watson |
||||||||
