CPU scaling benchmark
workers
2 +1 main
iters total
500M
166666666/stream
elapsed
1493.47 ms
total CPU used
4382.23 ms
speedup
2.93×
vs serial
efficiency
97.7%
of 3× ideal
| stream | spawn ms | spawned@ | work start@ | work end@ | work ms | reap wait ms |
|---|---|---|---|---|---|---|
| 0 (main) | 0 | 3.31 | 3.32 | 1447.43 | 1444.11 | 0 |
| 1 | 1.878 | 1.89 | 14.32 | 1490.48 | 1476.16 | 43.21 |
| 2 | 1.384 | 3.3 | 15.17 | 1477.13 | 1461.96 | 29.89 |
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.