да
значит какие-то ядра не догружены.
Сейчас при потоке ~200 мбит картина у нас выглядит следующим образом:
last pid: 26742; load averages: 0.18, 0.20, 0.22 up 53+00:15:47 12:40:44
135 processes: 10 running, 79 sleeping, 4 zombie, 42 waiting
CPU 0: 0.8% user, 0.0% nice, 0.4% system, 18.8% interrupt, 80.1% idle
CPU 1: 0.4% user, 0.0% nice, 0.8% system, 16.5% interrupt, 82.3% idle
CPU 2: 2.3% user, 0.0% nice, 0.4% system, 18.0% interrupt, 79.3% idle
CPU 3: 4.1% user, 0.0% nice, 0.0% system, 12.7% interrupt, 83.1% idle
CPU 4: 3.4% user, 0.0% nice, 0.4% system, 13.9% interrupt, 82.3% idle
CPU 5: 0.8% user, 0.0% nice, 0.0% system, 14.7% interrupt, 84.6% idle
CPU 6: 1.9% user, 0.0% nice, 0.0% system, 12.7% interrupt, 85.4% idle
CPU 7: 1.1% user, 0.0% nice, 0.0% system, 16.2% interrupt, 82.7% idle
Mem: 706M Active, 851M Inact, 283M Wired, 58M Cache, 112M Buf, 56M Free
Swap: 3973M Total, 3973M Free
PID USERNAME PRI NICE SIZE RES STATE C TIME WCPU COMMAND
11 root 171 ki31 0K 64K CPU3 3 1032.9 88.38% {idle: cpu3}
11 root 171 ki31 0K 64K RUN 6 1037.3 87.79% {idle: cpu6}
11 root 171 ki31 0K 64K CPU5 5 1039.9 87.60% {idle: cpu5}
11 root 171 ki31 0K 64K CPU0 0 977.5H 86.57% {idle: cpu0}
11 root 171 ki31 0K 64K CPU1 1 1037.4 83.40% {idle: cpu1}
11 root 171 ki31 0K 64K CPU4 4 1027.0 81.40% {idle: cpu4}
11 root 171 ki31 0K 64K CPU7 7 1046.1 80.47% {idle: cpu7}
11 root 171 ki31 0K 64K RUN 2 1020.8 79.98% {idle: cpu2}
12 root -68 - 0K 344K WAIT 7 107.4H 12.99% {irq272: igb1:que}
12 root -68 - 0K 344K WAIT 2 107.9H 12.16% {irq267: igb1:que}
12 root -68 - 0K 344K WAIT 0 102.5H 11.57% {irq265: igb1:que}
12 root -68 - 0K 344K WAIT 3 84.6H 10.89% {irq259: igb0:que}
12 root -68 - 0K 344K WAIT 5 108.2H 9.86% {irq270: igb1:que}
12 root -68 - 0K 344K WAIT 1 107.7H 8.79% {irq266: igb1:que}
12 root -68 - 0K 344K WAIT 5 83.0H 8.15% {irq261: igb0:que}
12 root -68 - 0K 344K WAIT 1 84.2H 7.86% {irq257: igb0:que}
12 root -68 - 0K 344K WAIT 4 108.5H 7.76% {irq269: igb1:que}
12 root -68 - 0K 344K WAIT 4 84.5H 7.67% {irq260: igb0:que}
12 root -68 - 0K 344K WAIT 2 82.5H 7.67% {irq258: igb0:que}
12 root -68 - 0K 344K CPU6 6 108.3H 7.18% {irq271: igb1:que}
934 nobody 48 0 4500K 2420K select 4 93.9H 7.18% softflowd
12 root -68 - 0K 344K WAIT 7 82.9H 5.66% {irq263: igb0:que}
12 root -68 - 0K 344K WAIT 0 79.8H 5.66% {irq256: last pid: 26742; load averages: 0.18, 0.20, 0.22 up 53+00:15:47 12:40:44
135 processes: 10 running, 79 sleeping, 4 zombie, 42 waiting
CPU 0: 0.8% user, 0.0% nice, 0.4% system, 18.8% interrupt, 80.1% idle
CPU 1: 0.4% user, 0.0% nice, 0.8% system, 16.5% interrupt, 82.3% idle
CPU 2: 2.3% user, 0.0% nice, 0.4% system, 18.0% interrupt, 79.3% idle
CPU 3: 4.1% user, 0.0% nice, 0.0% system, 12.7% interrupt, 83.1% idle
CPU 4: 3.4% user, 0.0% nice, 0.4% system, 13.9% interrupt, 82.3% idle
CPU 5: 0.8% user, 0.0% nice, 0.0% system, 14.7% interrupt, 84.6% idle
CPU 6: 1.9% user, 0.0% nice, 0.0% system, 12.7% interrupt, 85.4% idle
CPU 7: 1.1% user, 0.0% nice, 0.0% system, 16.2% interrupt, 82.7% idle
Mem: 706M Active, 851M Inact, 283M Wired, 58M Cache, 112M Buf, 56M Free
Swap: 3973M Total, 3973M Free
PID USERNAME PRI NICE SIZE RES STATE C TIME WCPU COMMAND
11 root 171 ki31 0K 64K CPU3 3 1032.9 88.38% {idle: cpu3}
11 root 171 ki31 0K 64K RUN 6 1037.3 87.79% {idle: cpu6}
11 root 171 ki31 0K 64K CPU5 5 1039.9 87.60% {idle: cpu5}
11 root 171 ki31 0K 64K CPU0 0 977.5H 86.57% {idle: cpu0}
11 root 171 ki31 0K 64K CPU1 1 1037.4 83.40% {idle: cpu1}
11 root 171 ki31 0K 64K CPU4 4 1027.0 81.40% {idle: cpu4}
11 root 171 ki31 0K 64K CPU7 7 1046.1 80.47% {idle: cpu7}
11 root 171 ki31 0K 64K RUN 2 1020.8 79.98% {idle: cpu2}
12 root -68 - 0K 344K WAIT 7 107.4H 12.99% {irq272: igb1:que}
12 root -68 - 0K 344K WAIT 2 107.9H 12.16% {irq267: igb1:que}
12 root -68 - 0K 344K WAIT 0 102.5H 11.57% {irq265: igb1:que}
12 root -68 - 0K 344K WAIT 3 84.6H 10.89% {irq259: igb0:que}
12 root -68 - 0K 344K WAIT 5 108.2H 9.86% {irq270: igb1:que}
12 root -68 - 0K 344K WAIT 1 107.7H 8.79% {irq266: igb1:que}
12 root -68 - 0K 344K WAIT 5 83.0H 8.15% {irq261: igb0:que}
12 root -68 - 0K 344K WAIT 1 84.2H 7.86% {irq257: igb0:que}
12 root -68 - 0K 344K WAIT 4 108.5H 7.76% {irq269: igb1:que}
12 root -68 - 0K 344K WAIT 4 84.5H 7.67% {irq260: igb0:que}
12 root -68 - 0K 344K WAIT 2 82.5H 7.67% {irq258: igb0:que}
12 root -68 - 0K 344K CPU6 6 108.3H 7.18% {irq271: igb1:que}
934 nobody 48 0 4500K 2420K select 4 93.9H 7.18% softflowd
12 root -68 - 0K 344K WAIT 7 82.9H 5.66% {irq263: igb0:que}
12 root -68 - 0K 344K WAIT 0 79.8H 5.66% {irq256: igb0:que}
12 root -68 - 0K 344K WAIT 6 82.5H 5.08% {irq262: igb0:que}
12 root -68 - 0K 344K WAIT 3 109.7H 4.05% {irq268: igb1:que}
1427 root 45 0 185M 179M bpf 2 19.5H 1.66% bandwidthd
1428 root 45 0 250M 226M bpf 7 59.9H 1.07% bandwidthd
1426 root 45 0 170M 166M bpf 3 727:51 0.88% bandwidthd
1425 root 44 0 127M 123M bpf 3 661:45 0.59% bandwidthd
igb0:que}
12 root -68 - 0K 344K WAIT 6 82.5H 5.08% {irq262: igb0:que}
12 root -68 - 0K 344K WAIT 3 109.7H 4.05% {irq268: igb1:que}
1427 root 45 0 185M 179M bpf 2 19.5H 1.66% bandwidthd
1428 root 45 0 250M 226M bpf 7 59.9H 1.07% bandwidthd
1426 root 45 0 170M 166M bpf 3 727:51 0.88% bandwidthd
1425 root 44 0 127M 123M bpf 3 661:45 0.59% bandwidthd
Соответственно cacti говорит о 19% утилизации CPU, что если смотреть эмпирически действительно соответствует загрузке всех ядер на 19-20%