CPU scaling benchmark
workers
2 +1 main
iters total
500M
166666666/stream
elapsed
1480.48 ms
total CPU used
4353.75 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 | 4.13 | 4.15 | 1446.1 | 1441.95 | 0 |
| 1 | 2.287 | 2.31 | 15.85 | 1477.59 | 1461.74 | 31.62 |
| 2 | 1.786 | 4.12 | 17.02 | 1467.08 | 1450.06 | 21.15 |
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.