Table 4 Mean execution times, speedups and parallel efficiency (Experiment 1 - query.fasta - 36 MB) - SparkBLAST vs CloudBLAST - Google Cloud
From: SparkBLAST: scalable BLAST processing using in-memory operations
# nodes | 1 | 2 | 4 | 8 | 16 | 32 | 64 |
|---|---|---|---|---|---|---|---|
SparkBLAST | |||||||
1 core | |||||||
Exec. time | 36,690.20 | 18,984.68 | 10,186.93 | 5,553.01 | 3,125.77 | 1,724.09 | 898.02 |
Speedup | 1 | 1.93 | 3.60 | 6.61 | 11.74 | 21.28 | 40.86 |
Efficiency | 1 | 0.97 | 0.90 | 0.83 | 0.73 | 0.67 | 0.64 |
SparkBLAST | |||||||
2 cores | |||||||
Exec. time | 28,983.44 | 14,532.37 | 7,943.95 | 4,291.57 | 2,246.37 | 1,260.17 | 693.30 |
Speedup | 1.00 | 1.99 | 3.65 | 6.75 | 12.90 | 23.00 | 41,81 |
Efficiency | 1.00 | 1.00 | 0.91 | 0.84 | 0.81 | 0.72 | 0.65 |
CloudBLAST | |||||||
Exec. time | 30,547.29 | 18,998.80 | 12,122.53 | 5,869.12 | 2,965.06 | 1,740.84 | 825.62 |
Speedup | 1.00 | 1.61 | 2.52 | 5.20 | 10.30 | 17.55 | 37.00 |
Efficiency | 1.00 | 0.80 | 0.63 | 0.65 | 0.64 | 0.55 | 0.58 |