An in-progress SQL OLAP database. There are two workloads, though the first one seems very abbreviated on my AMD run and not on the Intel run. The systemtime below also suggests this quick abort didn’t happen. Instead the first run looks single-threaded and the second runs a mix of single-threaded and once per core.
Topdown metrics with the timescales not matching the previous graph, also suggests this one could have quit the first workload early and mostly looking at the second workload where the retirement reaches highest point but also a blurring with frontend and backend stalls.
AMD metrics also show a shorter runtime so expect mostly the second workload. This has ~1/5 instructions are branches and less than 1/4 of the cores kept busy. A relatively low floating point.
elapsed 547.152
on_cpu 0.229 # 3.67 / 16 cores
utime 1829.973
stime 175.594
nvcsw 724734 # 95.49%
nivcsw 34194 # 4.51%
inblock 104 # 0.19/sec
onblock 66306048 # 121184.06/sec
cpu-clock 2006098320016 # 2006.098 seconds
task-clock 2006459295662 # 2006.459 seconds
page faults 43860238 # 21859.520/sec
context switches 761385 # 379.467/sec
cpu migrations 11771 # 5.867/sec
major page faults 2 # 0.001/sec
minor page faults 43860236 # 21859.519/sec
alignment faults 0 # 0.000/sec
emulation faults 0 # 0.000/sec
branches 2989521993930 # 209.081 branches per 1000 inst
branch misses 44666476992 # 1.49% branch miss
conditional 1855246304543 # 129.752 conditional branches per 1000 inst
indirect 336448903110 # 23.531 indirect branches per 1000 inst
cpu-cycles 8205492318167 # 0.99 GHz
instructions 14167935669400 # 1.73 IPC
slots 16457577130686 #
retiring 4848133219219 # 29.5% (36.1%)
-- ucode 12640610594 # 0.1%
-- fastpath 4835492608625 # 29.4%
frontend 4321707575747 # 26.3% (32.2%)
-- latency 2363442872946 # 14.4%
-- bandwidth 1958264702801 # 11.9%
backend 3675392035681 # 22.3% (27.4%)
-- cpu 521630866054 # 3.2%
-- memory 3153761169627 # 19.2%
speculation 580918774688 # 3.5% ( 4.3%)
-- branch mispredict 574059476463 # 3.5%
-- pipeline restart 6859298225 # 0.0%
smt-contention 3031336112693 # 18.4% ( 0.0%)
cpu-cycles 8220556664269 # 0.99 GHz
instructions 14166689794245 # 1.72 IPC
instructions 4726827381322 # 11.842 l2 access per 1000 inst
l2 hit from l1 36650616240 # 28.10% l2 miss
l2 miss from l1 5214331727 #
l2 hit from l2 pf 8812526969 #
l3 hit from l2 pf 3719577488 #
l3 miss from l2 pf 6793561232 #
instructions 4722831156232 # 20.931 float per 1000 inst
float 512 72 # 0.000 AVX-512 per 1000 inst
float 256 642 # 0.000 AVX-256 per 1000 inst
float 128 98853835720 # 20.931 AVX-128 per 1000 inst
float MMX 0 # 0.000 MMX per 1000 inst
float scalar 0 # 0.000 scalar per 1000 inst
Intel metrics
elapsed 1826.836
on_cpu 0.282 # 4.51 / 16 cores
utime 7881.944
stime 364.632
nvcsw 2932996 # 94.39%
nivcsw 174235 # 5.61%
inblock 276128 # 151.15/sec
onblock 70939448 # 38831.86/sec
cpu-clock 8239226921074 # 8239.227 seconds
task-clock 8240397739741 # 8240.398 seconds
page faults 116908038 # 14187.184/sec
context switches 3116153 # 378.156/sec
cpu migrations 121550 # 14.751/sec
major page faults 1481 # 0.180/sec
minor page faults 116906557 # 14187.004/sec
alignment faults 0 # 0.000/sec
emulation faults 0 # 0.000/sec
branches 8685405904523 # 190.950 branches per 1000 inst
branch misses 129488118910 # 1.49% branch miss
conditional 8685406471659 # 190.950 conditional branches per 1000 inst
indirect 2611251130674 # 57.409 indirect branches per 1000 inst
slots 56189409146582 #
retiring 27383968130937 # 48.7% (48.7%)
-- ucode 2029604665216 # 3.6%
-- fastpath 25354363465721 # 45.1%
frontend 10622806556285 # 18.9% (18.9%)
-- latency 4873119794808 # 8.7%
-- bandwidth 5749686761477 # 10.2%
backend 11019320981325 # 19.6% (19.6%)
-- cpu 4663030301016 # 8.3%
-- memory 6356290680309 # 11.3%
speculation 7380984669753 # 13.1% (13.1%)
-- branch mispredict 7196351812556 # 12.8%
-- pipeline restart 184632857197 # 0.3%
smt-contention 0 # 0.0% ( 0.0%)
cpu-cycles 27125366791737 # 1.41 GHz
instructions 50101367608730 # 1.85 IPC
l2 access 634116960124 # 22.630 l2 access per 1000 inst
l2 miss 173861651765 # 27.42% l2 miss
Process overview shows many benchmark runner processes spending almost all the time.
15898 processes
15540 benchmark_runne 301174.39 33694.99
68 clinfo 16.21 6.98
38 vulkaninfo 0.57 1.71
6 glxinfo:gdrv0 0.16 0.03
6 glxinfo:gl0 0.16 0.03
2 glxinfo 0.08 0.01
2 glxinfo:cs0 0.08 0.01
2 glxinfo:disk$0 0.08 0.01
2 glxinfo:sh0 0.08 0.01
2 glxinfo:shlo0 0.08 0.01
6 clang 0.07 0.05
4 vulkani:disk$0 0.06 0.18
6 php 0.04 0.16
2 llvmpipe-0 0.03 0.09
2 llvmpipe-1 0.03 0.09
2 llvmpipe-10 0.03 0.09
2 llvmpipe-11 0.03 0.09
2 llvmpipe-12 0.03 0.09
2 llvmpipe-13 0.03 0.09
2 llvmpipe-14 0.03 0.09
2 llvmpipe-15 0.03 0.09
2 llvmpipe-2 0.03 0.09
2 llvmpipe-3 0.03 0.09
2 llvmpipe-4 0.03 0.09
2 llvmpipe-5 0.03 0.09
2 llvmpipe-6 0.03 0.09
2 llvmpipe-7 0.03 0.09
2 llvmpipe-8 0.03 0.09
2 llvmpipe-9 0.03 0.09
3 rocminfo 0.03 0.00
1 lspci 0.00 0.02
1 ps 0.00 0.01
84 sh 0.00 0.00
13 gcc 0.00 0.00
9 gsettings 0.00 0.00
8 duckdb 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 gmain 0.00 0.00
5 phoronix-test-s 0.00 0.00
2 cc 0.00 0.00
2 dconf worker 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 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 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