Table 2 Performance estimation of LR vs. in-order
From: An optimizing pipeline stall reduction algorithm for power and performance on multi-core CPUs
Metrics | LR | In-order | Tomasulo |
|---|---|---|---|
Instructions per cycle (IPC) | 0.2462 | 0.2442 | 0.7327 |
Clocks per instruction (CPI) | 4.061 | 4.09 | 1.3648 |
Simulation speed (inst/sec) | 785 | 606 | 98021.9 |
Total number of instructions executed | 40516 | 41004 | 2350639 |
Time/Program | 164535.476 | 167706.36 | 3207982.549 |