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
1274.34 ms
total CPU used
5200.67 ms
speedup
4.08×
vs serial
efficiency
81.6%
of 5× ideal
stream spawn ms spawned@ work start@ work end@ work ms reap wait ms
0 (main) 0 8.23 8.24 875.85 867.61 0
1 2.073 2.09 14.23 1029.72 1015.49 154.01
2 1.63 3.74 16.28 996.55 980.27 126.25
3 2.557 6.32 35.34 1122.96 1087.62 247.24
4 1.885 8.22 21.75 1271.43 1249.68 395.74
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.