CPU scaling benchmark
workers
3 +1 main
iters total
500M
125000000/stream
elapsed
1125.99 ms
total CPU used
4379.65 ms
speedup
3.89×
vs serial
efficiency
97.3%
of 4× ideal
| stream | spawn ms | spawned@ | work start@ | work end@ | work ms | reap wait ms |
|---|---|---|---|---|---|---|
| 0 (main) | 0 | 5.24 | 5.25 | 1086.26 | 1081.01 | 0 |
| 1 | 2.019 | 2.03 | 15.63 | 1123.35 | 1107.72 | 37.18 |
| 2 | 1.552 | 3.61 | 15.62 | 1117.74 | 1102.12 | 31.59 |
| 3 | 1.605 | 5.24 | 16.96 | 1105.76 | 1088.8 | 19.64 |
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.