A python library which uses tensorflow and measures various AI models. One test that reports both training and inference scores. The number of running processes seems to bounce to the number of cores and single threaded and then in between. Looks like it also varies with sub workloads.
Topdown metrics are similar to the “tensorflow” test in being heavily backend bound. That test has more regular patterns while this one goes to different sub tests.
AMD metrics show not much floating point code and a high backend stalls and low retiring and frontend stalls. This report is one of the first with opcache, tlb and icache statistics.
elapsed 1051.310
on_cpu 0.735 # 11.76 / 16 cores
utime 12130.094
stime 232.507
nvcsw 15457321 # 97.25%
nivcsw 436597 # 2.75%
inblock 0 # 0.00/sec
onblock 12584 # 11.97/sec
cpu-clock 12368006983271 # 12368.007 seconds
task-clock 12373062836640 # 12373.063 seconds
page faults 26680062 # 2156.302/sec
context switches 15898993 # 1284.968/sec
cpu migrations 3022862 # 244.310/sec
major page faults 7 # 0.001/sec
minor page faults 26680055 # 2156.302/sec
alignment faults 0 # 0.000/sec
emulation faults 0 # 0.000/sec
branches 957490729882 # 31.389 branches per 1000 inst
branch misses 13287236219 # 1.39% branch miss
conditional 801168210108 # 26.264 conditional branches per 1000 inst
indirect 23166096714 # 0.759 indirect branches per 1000 inst
cpu-cycles 49761243136903 # 2.95 GHz
instructions 30637757626442 # 0.62 IPC low
slots 99453847169220 #
retiring 10179875202861 # 10.2% (12.1%) low
-- ucode 2580640907 # 0.0%
-- fastpath 10177294561954 # 10.2%
frontend 3462384584201 # 3.5% ( 4.1%) low
-- latency 2663176605084 # 2.7%
-- bandwidth 799207979117 # 0.8%
backend 70000221724395 # 70.4% (83.5%) high
-- cpu 36281460856251 # 36.5%
-- memory 33718760868144 # 33.9%
speculation 141920805302 # 0.1% ( 0.2%) low
-- branch mispredict 124024703083 # 0.1%
-- pipeline restart 17896102219 # 0.0%
smt-contention 15668164812391 # 15.8% ( 0.0%)
cpu-cycles 49849520997432 # 2.96 GHz
instructions 30646476066015 # 0.61 IPC low
instructions 10204482310837 # 131.096 l2 access per 1000 inst
l2 hit from l1 992671755236 # 12.84% l2 miss
l2 miss from l1 56974474740 #
l2 hit from l2 pf 230317005078 #
l3 hit from l2 pf 74512435238 #
l3 miss from l2 pf 40261901613 #
instructions 10210314170078 # 31.654 float per 1000 inst
float 512 68 # 0.000 AVX-512 per 1000 inst
float 256 15380490758 # 1.506 AVX-256 per 1000 inst
float 128 307817367452 # 30.148 AVX-128 per 1000 inst
float MMX 0 # 0.000 MMX per 1000 inst
float scalar 6 # 0.000 scalar per 1000 inst
instructions 2668409 #
opcache 996007 # 373.259 opcache per 1000 inst
opcache miss 537203 # 53.9% opcache miss rate
l1 dTLB miss 4865 # 1.823 L1 dTLB per 1000 inst
l2 dTLB miss 990 # 0.371 L2 dTLB per 1000 inst
instructions 2715789 #
icache 1316573 # 484.785 icache per 1000 inst
icache miss 112561 # 8.5% icache miss rate
l1 iTLB miss 13 # 0.005 L1 iTLB per 1000 inst
l2 iTLB miss 0 # 0.000 L2 iTLB per 1000 inst
tlb flush 18 # 0.007 TLB flush per 1000 inst
Intel metrics
elapsed 1337.941
on_cpu 0.494 # 7.91 / 16 cores
utime 10366.642
stime 215.507
nvcsw 8160500 # 93.34%
nivcsw 582164 # 6.66%
inblock 148904 # 111.29/sec
onblock 1344 # 1.00/sec
cpu-clock 10557710888684 # 10557.711 seconds
task-clock 10563300378839 # 10563.300 seconds
page faults 22965931 # 2174.125/sec
context switches 8749217 # 828.265/sec
cpu migrations 2136601 # 202.266/sec
major page faults 724 # 0.069/sec
minor page faults 22965207 # 2174.056/sec
alignment faults 0 # 0.000/sec
emulation faults 0 # 0.000/sec
branches 1798252958796 # 41.581 branches per 1000 inst
branch misses 8481393140 # 0.47% branch miss
conditional 1798252972524 # 41.581 conditional branches per 1000 inst
indirect 548815196792 # 12.690 indirect branches per 1000 inst
slots 58533044377064 #
retiring 23888560940405 # 40.8% (40.8%)
-- ucode 1622686986715 # 2.8%
-- fastpath 22265873953690 # 38.0%
frontend 7616630872137 # 13.0% (13.0%)
-- latency 5497768921797 # 9.4%
-- bandwidth 2118861950340 # 3.6%
backend 25098776551734 # 42.9% (42.9%)
-- cpu 11916166002514 # 20.4%
-- memory 13182610549220 # 22.5%
speculation 2054328069108 # 3.5% ( 3.5%)
-- branch mispredict 1869991804975 # 3.2%
-- pipeline restart 184336264133 # 0.3%
smt-contention 0 # 0.0% ( 0.0%)
cpu-cycles 31128010635019 # 1.46 GHz
instructions 57746411381473 # 1.86 IPC
l2 access 1262468579018 # 48.640 l2 access per 1000 inst
l2 miss 297475919726 # 23.56% l2 miss
Process overview shows mostly invocations of python
410 processes
54 python3 600264.00 9015.82
68 clinfo 17.19 5.00
38 vulkaninfo 1.15 0.96
4 vulkani:disk$0 0.13 0.11
6 php 0.11 0.23
6 glxinfo:gdrv0 0.08 0.09
6 glxinfo:gl0 0.08 0.09
2 llvmpipe-0 0.07 0.06
2 llvmpipe-1 0.07 0.06
2 llvmpipe-10 0.07 0.06
2 llvmpipe-11 0.07 0.06
2 llvmpipe-12 0.07 0.06
2 llvmpipe-13 0.07 0.06
2 llvmpipe-14 0.07 0.06
2 llvmpipe-15 0.07 0.06
2 llvmpipe-2 0.07 0.06
2 llvmpipe-3 0.07 0.06
2 llvmpipe-4 0.07 0.06
2 llvmpipe-5 0.07 0.06
2 llvmpipe-6 0.07 0.06
2 llvmpipe-7 0.07 0.06
2 llvmpipe-8 0.07 0.06
2 llvmpipe-9 0.07 0.06
2 glxinfo 0.05 0.04
6 clang 0.04 0.08
2 glxinfo:cs0 0.04 0.04
2 glxinfo:disk$0 0.04 0.04
2 glxinfo:sh0 0.04 0.04
2 glxinfo:shlo0 0.04 0.04
3 rocminfo 0.03 0.00
1 lspci 0.00 0.02
81 sh 0.00 0.00
12 gcc 0.00 0.00
11 gsettings 0.00 0.00
8 stat 0.00 0.00
8 systemd-detect- 0.00 0.00
6 llvm-link 0.00 0.00
5 phoronix-test-s 0.00 0.00
4 uname 0.00 0.00
3 gmain 0.00 0.00
3 lscpu 0.00 0.00
2 dconf worker 0.00 0.00
2 dmesg 0.00 0.00
2 file 0.00 0.00
2 which 0.00 0.00
2 xset 0.00 0.00
1 ai-benchmark 0.00 0.00
1 cat 0.00 0.00
1 cc 0.00 0.00
1 date 0.00 0.00
1 dirname 0.00 0.00
1 dmidecode 0.00 0.00
1 grep 0.00 0.00
1 ifconfig 0.00 0.00
1 ip 0.00 0.00
1 lsmod 0.00 0.00
1 mktemp 0.00 0.00
1 ps 0.00 0.00
1 python 0.00 0.00
1 qdbus 0.00 0.00
1 readlink 0.00 0.00
1 realpath 0.00 0.00
1 sed 0.00 0.00
1 sort 0.00 0.00
1 stty 0.00 0.00
1 sysctl 0.00 0.00
1 systemctl 0.00 0.00
1 template.sh 0.00 0.00
1 wc 0.00 0.00
1 xrandr 0.00 0.00
1 processes running
61 maximum processes
An example computation block
206953) ai-benchmark cpu=12 start=5.67 finish=1046.64
206954) python3 cpu=2 start=5.68 finish=1046.37
206955) python3 cpu=15 start=5.86 finish=1046.37
206956) python3 cpu=0 start=5.86 finish=1046.37
206957) python3 cpu=1 start=5.86 finish=1046.37
206958) python3 cpu=10 start=5.86 finish=1046.37
206959) python3 cpu=11 start=5.86 finish=1046.37
206960) python3 cpu=13 start=5.86 finish=1046.37
206961) python3 cpu=12 start=5.86 finish=1046.37
206962) python3 cpu=14 start=5.86 finish=1046.37
206963) python3 cpu=7 start=5.86 finish=1046.37
206964) python3 cpu=8 start=5.86 finish=1046.37
206965) python3 cpu=9 start=5.86 finish=1046.37
206966) python3 cpu=6 start=5.86 finish=1046.37
206967) python3 cpu=3 start=5.86 finish=1046.37
206968) python3 cpu=5 start=5.86 finish=1046.37
206969) python3 cpu=4 start=5.86 finish=1046.37
206970) file cpu=7 start=6.72 finish=6.73
206971) uname cpu=1 start=6.73 finish=6.73
206972) python3 cpu=15 start=6.73 finish=7.80
206973) file cpu=1 start=6.75 finish=6.76
206974) uname cpu=10 start=6.76 finish=6.76
206975) cat cpu=4 start=6.76 finish=6.76
206976) lscpu cpu=1 start=6.76 finish=6.77
206977) sysctl cpu=10 start=6.77 finish=6.77
206978) dmesg cpu=4 start=6.77 finish=6.78
206979) python3 cpu=1 start=6.78 finish=7.79
206981) ?? cpu=0 start=7.79 finish=0.00
206982) python3 cpu=13 start=7.80 finish=1046.37
206983) python3 cpu=0 start=7.80 finish=1046.37
206984) python3 cpu=11 start=7.80 finish=1046.37
206985) python3 cpu=9 start=7.80 finish=1046.37
206986) python3 cpu=9 start=7.80 finish=1046.37
206987) python3 cpu=8 start=7.80 finish=1046.37
206988) python3 cpu=15 start=7.80 finish=1046.37
206989) python3 cpu=6 start=7.80 finish=1046.37
206990) python3 cpu=13 start=7.80 finish=1046.37
206991) python3 cpu=5 start=7.80 finish=1046.37
206992) python3 cpu=10 start=7.80 finish=1046.37
206993) python3 cpu=14 start=7.80 finish=1046.37
206994) python3 cpu=12 start=7.80 finish=1046.37
206995) python3 cpu=3 start=7.80 finish=1046.37
206996) python3 cpu=4 start=7.80 finish=1046.37
206997) python3 cpu=12 start=7.81 finish=1046.37
206998) python3 cpu=11 start=7.81 finish=7.81
207000) python3 cpu=7 start=14.34 finish=1046.37
207001) python3 cpu=7 start=14.34 finish=1046.37
207002) python3 cpu=5 start=14.34 finish=1046.37
207003) python3 cpu=1 start=14.34 finish=1046.37
207004) python3 cpu=1 start=14.34 finish=1046.37
207005) python3 cpu=1 start=14.34 finish=1046.37
207006) python3 cpu=2 start=14.34 finish=1046.37
207007) python3 cpu=4 start=14.34 finish=1046.37
207008) python3 cpu=6 start=14.34 finish=1046.37
207009) python3 cpu=8 start=14.34 finish=1046.37
207010) python3 cpu=3 start=14.34 finish=1046.37
207011) python3 cpu=10 start=14.35 finish=1046.37
207012) python3 cpu=14 start=14.35 finish=1046.37
207013) python3 cpu=14 start=14.35 finish=1046.37
207014) python3 cpu=13 start=14.35 finish=1046.37
207015) python3 cpu=12 start=14.35 finish=1046.37
207016) python3 cpu=2 start=14.35 finish=1046.37
207017) python3 cpu=7 start=14.70 finish=14.70