A test that generates workloads resembling web server workloads. Quick running but many threads.
Topdown profile shows higher levels of both backend stalls and frontend stalls.
AMD metrics show a large amount of L2 access? Not much floating point and a high opcache miss rate.
elapsed 76.014
on_cpu 0.790 # 12.64 / 16 cores
utime 960.128
stime 0.969
nvcsw 2441 # 3.57%
nivcsw 65967 # 96.43%
inblock 0 # 0.00/sec
onblock 12568 # 165.34/sec
cpu-clock 961212794580 # 961.213 seconds
task-clock 961219693426 # 961.220 seconds
page faults 174119 # 181.144/sec
context switches 68613 # 71.381/sec
cpu migrations 794 # 0.826/sec
major page faults 43 # 0.045/sec
minor page faults 174076 # 181.099/sec
alignment faults 0 # 0.000/sec
emulation faults 0 # 0.000/sec
branches 8490158192 # 233.934 branches per 1000 inst
branch misses 122343044 # 1.44% branch miss
conditional 6245003590 # 172.072 conditional branches per 1000 inst
indirect 174774988 # 4.816 indirect branches per 1000 inst
cpu-cycles 3619643263703 # 2.97 GHz
instructions 36257254806 # 0.01 IPC low
slots 7241048093232 #
retiring 481361243242 # 6.6% ( 7.1%) low
-- ucode 56944409 # 0.0%
-- fastpath 481304298833 # 6.6%
frontend 2366660898335 # 32.7% (35.1%)
-- latency 70688934948 # 1.0%
-- bandwidth 2295971963387 # 31.7%
backend 3883706267866 # 53.6% (57.7%)
-- cpu 149362501021 # 2.1%
-- memory 3734343766845 # 51.6%
speculation 2804700251 # 0.0% ( 0.0%) low
-- branch mispredict 2049764610 # 0.0%
-- pipeline restart 754935641 # 0.0%
smt-contention 506509742595 # 7.0% ( 0.0%)
cpu-cycles 3614193928091 # 2.96 GHz
instructions 36147971263 # 0.01 IPC low
instructions 11965344127 # 42861.333 l2 access per 1000 inst
l2 hit from l1 228293533402 # 26.95% l2 miss
l2 miss from l1 20386259467 #
l2 hit from l2 pf 166706468197 #
l3 hit from l2 pf 117806766518 #
l3 miss from l2 pf 43833622 #
instructions 12235477579 # 18.695 float per 1000 inst
float 512 66 # 0.000 AVX-512 per 1000 inst
float 256 596 # 0.000 AVX-256 per 1000 inst
float 128 228741691 # 18.695 AVX-128 per 1000 inst
float MMX 0 # 0.000 MMX per 1000 inst
float scalar 0 # 0.000 scalar per 1000 inst
instructions 36020965289 #
opcache 9790344647 # 271.796 opcache per 1000 inst
opcache miss 4930331972 # 50.4% opcache miss rate
l1 dTLB miss 11689841338 # 324.529 L1 dTLB per 1000 inst
l2 dTLB miss 26979105 # 0.749 L2 dTLB per 1000 inst
instructions 36020505704 #
icache 6350080470 # 176.291 icache per 1000 inst
icache miss 2161166540 # 34.0% icache miss rate
l1 iTLB miss 9196328 # 0.255 L1 iTLB per 1000 inst
l2 iTLB miss 0 # 0.000 L2 iTLB per 1000 inst
tlb flush 19035 # 0.001 TLB flush per 1000 inst
Intel metrics also show a high amount of L2 access and a very large amount of backend memory stalls.
elapsed 76.197
on_cpu 0.788 # 12.61 / 16 cores
utime 960.003
stime 0.625
nvcsw 2219 # 3.37%
nivcsw 63631 # 96.63%
inblock 2216 # 29.08/sec
onblock 1320 # 17.32/sec
cpu-clock 960779969649 # 960.780 seconds
task-clock 960785636136 # 960.786 seconds
page faults 162859 # 169.506/sec
context switches 66055 # 68.751/sec
cpu migrations 855 # 0.890/sec
major page faults 24 # 0.025/sec
minor page faults 162835 # 169.481/sec
alignment faults 0 # 0.000/sec
emulation faults 0 # 0.000/sec
branches 5326251888 # 224.184 branches per 1000 inst
branch misses 21464693 # 0.40% branch miss
conditional 5326266800 # 224.185 conditional branches per 1000 inst
indirect 656377561 # 27.627 indirect branches per 1000 inst
slots 5160889464836 #
retiring 291731107053 # 5.7% ( 5.7%) low
-- ucode 100764509180 # 2.0%
-- fastpath 190966597873 # 3.7%
frontend 346184570707 # 6.7% ( 6.7%)
-- latency 337244658529 # 6.5%
-- bandwidth 8939912178 # 0.2%
backend 4521787284195 # 87.6% (87.6%) high
-- cpu 572017397254 # 11.1%
-- memory 3949769886941 # 76.5%
speculation 3292312798 # 0.1% ( 0.1%) low
-- branch mispredict 3103520615 # 0.1%
-- pipeline restart 188792183 # 0.0%
smt-contention 0 # 0.0% ( 0.0%)
cpu-cycles 3597502784460 # 1.82 GHz
instructions 32690767154 # 0.01 IPC low
l2 access 547037131507 # 21244.104 l2 access per 1000 inst
l2 miss 212437512662 # 38.83% l2 miss
cpu-cycles 5628197624978 # 85.4% memory latency
load stalls 4804943018686 # 24.1% l1 bound
l1 miss 3448039312113 # 38.0% l2 bound
l2 miss 1310663434850 # 23.3% l3 bound
l3 miss 956314300 # 0.0% dram bound
store_stalls 3981547654 # 0.1% store bound
Process summary shows time in the ebizzy application.
824 processes
476 ebizzy 147756.72 17.82
68 clinfo 14.88 8.31
38 vulkaninfo 1.33 1.14
4 vulkani:disk$0 0.15 0.12
6 php 0.09 0.14
2 llvmpipe-0 0.08 0.06
2 llvmpipe-1 0.08 0.06
2 llvmpipe-10 0.08 0.06
2 llvmpipe-11 0.08 0.06
2 llvmpipe-12 0.08 0.06
2 llvmpipe-13 0.08 0.06
2 llvmpipe-14 0.08 0.06
2 llvmpipe-15 0.08 0.06
2 llvmpipe-2 0.08 0.06
2 llvmpipe-3 0.08 0.06
2 llvmpipe-4 0.08 0.06
2 llvmpipe-5 0.08 0.06
2 llvmpipe-6 0.08 0.06
2 llvmpipe-7 0.08 0.06
2 llvmpipe-8 0.08 0.06
2 llvmpipe-9 0.08 0.06
6 glxinfo:gdrv0 0.07 0.11
6 glxinfo:gl0 0.07 0.11
6 clang 0.06 0.06
2 glxinfo 0.05 0.04
2 glxinfo:cs0 0.05 0.04
2 glxinfo:disk$0 0.05 0.04
2 glxinfo:sh0 0.05 0.04
2 glxinfo:shlo0 0.05 0.04
1 lspci 0.00 0.02
82 sh 0.00 0.00
13 gcc 0.00 0.00
13 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
3 rocminfo 0.00 0.00
2 cc 0.00 0.00
2 gmain 0.00 0.00
2 lscpu 0.00 0.00
2 uname 0.00 0.00
2 which 0.00 0.00
2 xset 0.00 0.00
1 date 0.00 0.00
1 dconf worker 0.00 0.00
1 dirname 0.00 0.00
1 dmesg 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 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 systemctl 0.00 0.00
1 template.sh 0.00 0.00
1 wc 0.00 0.00
1 xrandr 0.00 0.00
0 processes running
47 maximum processes
Computation blocks look as follows
958113) ebizzy cpu=6 start=5.64 finish=25.65
958114) ebizzy cpu=15 start=5.64 finish=25.65
958115) ebizzy cpu=9 start=5.64 finish=25.65
958116) ebizzy cpu=12 start=5.64 finish=25.65
958117) ebizzy cpu=2 start=5.64 finish=25.65
958118) ebizzy cpu=0 start=5.64 finish=25.65
958119) ebizzy cpu=7 start=5.64 finish=25.65
958120) ebizzy cpu=0 start=5.64 finish=25.65
958121) ebizzy cpu=4 start=5.64 finish=25.65
958122) ebizzy cpu=2 start=5.64 finish=25.64
958123) ebizzy cpu=4 start=5.64 finish=25.64
958124) ebizzy cpu=10 start=5.64 finish=25.64
958125) ebizzy cpu=12 start=5.64 finish=25.64
958126) ebizzy cpu=13 start=5.64 finish=25.64
958127) ebizzy cpu=8 start=5.64 finish=25.64
958128) ebizzy cpu=11 start=5.64 finish=25.64
958129) ebizzy cpu=13 start=5.64 finish=25.64
958130) ebizzy cpu=1 start=5.64 finish=25.64
958131) ebizzy cpu=4 start=5.64 finish=25.64
958132) ebizzy cpu=15 start=5.64 finish=25.64
958133) ebizzy cpu=3 start=5.64 finish=25.64
958134) ebizzy cpu=14 start=5.64 finish=25.64
958135) ebizzy cpu=3 start=5.64 finish=25.64
958136) ebizzy cpu=10 start=5.64 finish=25.64
958137) ebizzy cpu=11 start=5.64 finish=25.64
958138) ebizzy cpu=6 start=5.64 finish=25.64
958139) ebizzy cpu=1 start=5.64 finish=25.64
958140) ebizzy cpu=6 start=5.64 finish=25.64
958141) ebizzy cpu=5 start=5.64 finish=25.64
958142) ebizzy cpu=15 start=5.64 finish=25.64
958143) ebizzy cpu=9 start=5.64 finish=25.64
958144) ebizzy cpu=14 start=5.64 finish=25.64
958145) ebizzy cpu=8 start=5.64 finish=25.64
958146) ebizzy cpu=7 start=5.64 finish=25.64