CPU scaling benchmark
workers
2 +1 main
iters total
500M
166666666/stream
elapsed
1473.07 ms
total CPU used
4337.75 ms
speedup
2.94×
vs serial
efficiency
98%
of 3× ideal
| stream | spawn ms | spawned@ | work start@ | work end@ | work ms | reap wait ms |
|---|---|---|---|---|---|---|
| 0 (main) | 0 | 3.33 | 3.33 | 1441.54 | 1438.21 | 0 |
| 1 | 1.789 | 1.8 | 16.04 | 1462.75 | 1446.71 | 21.37 |
| 2 | 1.482 | 3.31 | 17.53 | 1470.36 | 1452.83 | 28.94 |
main
w1
w2
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.