CPU scaling benchmark
workers
3 +1 main
iters total
500M
125000000/stream
elapsed
1111.57 ms
total CPU used
4348.59 ms
speedup
3.91×
vs serial
efficiency
97.8%
of 4× ideal
| stream | spawn ms | spawned@ | work start@ | work end@ | work ms | reap wait ms |
|---|---|---|---|---|---|---|
| 0 (main) | 0 | 5.38 | 5.39 | 1089.13 | 1083.74 | 0 |
| 1 | 2.153 | 2.17 | 16.68 | 1107.96 | 1091.28 | 19.51 |
| 2 | 1.584 | 3.78 | 17.17 | 1103.11 | 1085.94 | 14.11 |
| 3 | 1.576 | 5.37 | 17.24 | 1104.87 | 1087.63 | 22.38 |
main
w1
w2
w3
fork+handshake
CPU work
parent reap wait
what this measures
Each stream runs a tight integer LCG loop — working set is one CPU register, no memory access,
no shared data. Speedup = sum(stream CPU time) / wall-clock elapsed. Efficiency = speedup / (workers+1).
100% efficiency means perfect linear scaling; less than 100% is the cost of serial fork setup,
reap tail, SMT/core contention.