CPU scaling benchmark
workers
3 +1 main
iters total
500M
125000000/stream
elapsed
1113.79 ms
total CPU used
4351.78 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 | 4.8 | 4.81 | 1085.52 | 1080.71 | 0 |
| 1 | 1.883 | 1.89 | 16.75 | 1107.57 | 1090.82 | 23.92 |
| 2 | 1.453 | 3.37 | 17.2 | 1104.39 | 1087.19 | 19.01 |
| 3 | 1.398 | 4.79 | 18.61 | 1111.67 | 1093.06 | 26.82 |
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.