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
6 +1 main
iters total
500M
71428571/stream
elapsed
1225.56 ms
total CPU used
7070.6 ms
speedup
5.77×
vs serial
efficiency
82.4%
of 7× ideal
stream spawn ms spawned@ work start@ work end@ work ms reap wait ms
0 (main) 0 17.26 17.27 1153.77 1136.5 0
1 2.174 2.19 15.89 647.75 631.86 0.15
2 1.557 3.77 16.14 1032.03 1015.89 0.22
3 2.003 5.78 17.97 995.41 977.44 0.23
4 2.442 8.24 60.96 1164.43 1103.47 10.78
5 7.194 15.45 48.81 1222.66 1173.85 69.03
6 1.784 17.25 43.43 1075.02 1031.59 0.25
main
w1
w2
w3
w4
w5
w6
    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.