CPU scaling benchmark

W=0 W=1 W=2 W=3 W=4 W=6 W=8 W=9 W=10 W=12 W=16 iters: 100M 500M 1B 2B
workers
4 +1 main
iters total
500M
100000000/stream
elapsed
1146.31 ms
total CPU used
5341.99 ms
speedup
4.66×
vs serial
efficiency
93.2%
of 5× ideal
stream spawn ms spawned@ work start@ work end@ work ms reap wait ms
0 (main) 0 6.54 6.56 1086.18 1079.62 0
1 1.875 1.89 16.96 1055.99 1039.03 0.1
2 1.464 3.39 26.08 1142.94 1116.86 59.43
3 1.551 4.95 20.17 1022.16 1001.99 0.16
4 1.556 6.53 38.02 1142.51 1104.49 56.46
main
w1
w2
w3
w4
    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.