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
8 +1 main
iters total
500M
55555555/stream
elapsed
1164.16 ms
total CPU used
8914.82 ms
speedup
7.66×
vs serial
efficiency
85.1%
of 9× ideal
stream spawn ms spawned@ work start@ work end@ work ms reap wait ms
0 (main) 0 18.36 18.38 1154.47 1136.09 0
1 2.235 2.25 19.35 1051.73 1032.38 0.15
2 1.636 3.92 20.18 931.69 911.51 0.21
3 1.622 5.56 23.88 866.29 842.41 0.22
4 5.207 10.79 49.67 1161.2 1111.53 6.85
5 1.869 12.68 56.58 1145.94 1089.36 0.24
6 1.586 14.29 46.55 1002.6 956.05 0.27
7 1.599 15.9 56.64 1073.15 1016.51 0.28
8 2.433 18.35 62.68 881.66 818.98 0.3
main
w1
w2
w3
w4
w5
w6
w7
w8
    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.