Table 2.
Parallel implementation | Cores | Total Time | ODE time | PDE time | Speedup |
---|---|---|---|---|---|
Cluster | 1 | 546,507 | 523,331 | 23,177 | — |
Cluster | 64 | 8,934 | 8,313 | 607 | 61.2 |
| |||||
Multi-GPU | 64 + 16 GPUs | 1,302 | 682 | 611 | 420 |
Parallel implementation | Cores | Total Time | ODE time | PDE time | Speedup |
---|---|---|---|---|---|
Cluster | 1 | 546,507 | 523,331 | 23,177 | — |
Cluster | 64 | 8,934 | 8,313 | 607 | 61.2 |
| |||||
Multi-GPU | 64 + 16 GPUs | 1,302 | 682 | 611 | 420 |