1 | Largest communication load is communicating 16 words to be compared to calculating 16 updates -- each taking time tcalc |
2 | Each communication is one value of ? probably stored in a 4 byte word and takes time tcomm |
3 | Then on 16 processors, T16 = 16tcalc + 16tcomm |
4 | Speedup S = T1/T16 = 12.25 / (1 + tcomm/tcalc) |
5 | or S = 12.25 / (1 + 0.25 tcomm/tfloat) |
6 | or S ? 12.25 * (1 - 0.25 tcomm/tfloat) |