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
1134.77 ms
total CPU used
5352.89 ms
speedup
4.72×
vs serial
efficiency
94.4%
of 5× ideal
stream spawn ms spawned@ work start@ work end@ work ms reap wait ms
0 (main) 0 7.48 7.5 1091.9 1084.4 0
1 2.073 2.09 17.64 1089.6 1071.96 0.11
2 1.552 3.67 19.79 1015.89 996.1 3.2
3 1.623 5.31 20.58 1127.02 1106.44 35.25
4 2.133 7.47 38.02 1132.01 1093.99 40.23
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.