A test of a cloud-native distributed SQL database. Looks like a version from cockroachlabs is here. This runs on a local node trying different workloads. I picked 60% read and 256 concurrency. Looks like a variable number of runable processes that sometimes exceeds the available cores.

Topdown profile looks to be dominated by frontend stalls with a low retirement rate.

Topdown metrics show a little over half the cores busy. The IPC and retirement rate is low with frontend latency high. There is still a moderate L2 access rate and L2 miss rate

elapsed              334.639
on_cpu               0.597          # 9.56 / 16 cores
utime                2696.582
stime                502.219
nvcsw                18549411       # 76.74%
nivcsw               5620919        # 23.26%
inblock              0              # 0.00/sec
onblock              4620296        # 13806.79/sec
cpu-clock            4520258374771  # 4520.258 seconds
task-clock           4522943060763  # 4522.943 seconds
page faults          1432061        # 316.621/sec
context switches     27854844       # 6158.566/sec
cpu migrations       2728206        # 603.193/sec
major page faults    4              # 0.001/sec
minor page faults    1432057        # 316.621/sec
alignment faults     0              # 0.000/sec
emulation faults     0              # 0.000/sec
branches             1974030303012  # 201.912 branches per 1000 inst
branch misses        102407170177   # 5.19% branch miss
conditional          1284689450537  # 131.404 conditional branches per 1000 inst
indirect             40873144438    # 4.181 indirect branches per 1000 inst
cpu-cycles           19435892478859 # 3.63 GHz
instructions         9809598549648  # 0.50 IPC low
slots                38801128676808 #
retiring             3318868095039  #  8.6% ( 9.4%) low
-- ucode             9022406545     #     0.0%
-- fastpath          3309845688494  #     8.5%
frontend             22335021009203 # 57.6% (63.2%) high
-- latency           18761127031386 #    48.4%
-- bandwidth         3573893977817  #     9.2%
backend              9023155560220  # 23.3% (25.5%)
-- cpu               1160790156481  #     3.0%
-- memory            7862365403739  #    20.3%
speculation          658238820427   #  1.7% ( 1.9%)
-- branch mispredict 649146024898   #     1.7%
-- pipeline restart  9092795529     #     0.0%
smt-contention       3464824277011  #  8.9% ( 0.0%)
cpu-cycles           19420934695654 # 3.63 GHz
instructions         9799859808336  # 0.50 IPC low
instructions         3263133452367  # 116.177 l2 access per 1000 inst
l2 hit from l1       337971873728   # 46.20% l2 miss
l2 miss from l1      148712430714   #
l2 hit from l2 pf    14700493681    #
l3 hit from l2 pf    17772088522    #
l3 miss from l2 pf   8655223043     #
instructions         3257890384531  # 45.130 float per 1000 inst
float 512            51             # 0.000 AVX-512 per 1000 inst
float 256            594            # 0.000 AVX-256 per 1000 inst
float 128            147029392456   # 45.130 AVX-128 per 1000 inst
float MMX            0              # 0.000 MMX per 1000 inst
float scalar         0              # 0.000 scalar per 1000 inst
instructions         2397986        #
opcache              896597         # 373.896 opcache per 1000 inst
opcache miss         473659         # 52.8% opcache miss rate
l1 dTLB miss         4325           # 1.804 L1 dTLB per 1000 inst
l2 dTLB miss         1045           # 0.436 L2 dTLB per 1000 inst
instructions         2408738        #
icache               1178593        # 489.299 icache per 1000 inst
icache miss          107606         #  9.1% 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            19             # 0.008 TLB flush per 1000 inst

Intel metrics

elapsed              335.440
on_cpu               0.086          # 1.37 / 16 cores
utime                242.311
stime                217.011
nvcsw                15997404       # 98.96%
nivcsw               168694         # 1.04%
inblock              16             # 0.05/sec
onblock              1608           # 4.79/sec
cpu-clock            4503641628988  # 4503.642 seconds
task-clock           4505774179785  # 4505.774 seconds
page faults          1379992        # 306.272/sec
context switches     26579581       # 5899.004/sec
cpu migrations       3427600        # 760.713/sec
major page faults    0              # 0.000/sec
minor page faults    1379992        # 306.272/sec
alignment faults     0              # 0.000/sec
emulation faults     0              # 0.000/sec
branches             1675102884730  # 199.003 branches per 1000 inst
branch misses        57253165540    # 3.42% branch miss
conditional          1675102907834  # 199.003 conditional branches per 1000 inst
indirect             412498493591   # 49.005 indirect branches per 1000 inst
slots                21972129346064 #
retiring             5215647780921  # 23.7% (23.7%)
-- ucode             538637393473   #     2.5%
-- fastpath          4677010387448  #    21.3%
frontend             11430702200923 # 52.0% (52.0%) high
-- latency           8623027187045  #    39.2%
-- bandwidth         2807675013878  #    12.8%
backend              3256786565006  # 14.8% (14.8%) low
-- cpu               1180244257814  #     5.4%
-- memory            2076542307192  #     9.5%
speculation          2494472949221  # 11.4% (11.4%) high
-- branch mispredict 2423272684439  #    11.0%
-- pipeline restart  71200264782    #     0.3%
smt-contention       0              #  0.0% ( 0.0%)
cpu-cycles           14285580800826 # 2.67 GHz
instructions         9581793809048  # 0.67 IPC low
l2 access            586092773541   # 124.393 l2 access per 1000 inst
l2 miss              287650250348   # 49.08% l2 miss
cpu-cycles           7019364076935  # 44.0% memory latency
load stalls          3009622724986  #  3.4% l1 bound
l1 miss              2769044358629  # 11.7% l2 bound
l2 miss              1950347371896  # 13.5% l3 bound
l3 miss              1005857251457  # 14.3% dram bound
store_stalls         77504801801    #  1.1% store bound

Process overview shows the cockroach process with the most time.

632 processes
	381 cockroach            125949.13 34009.39
	 38 vulkaninfo               1.50     1.14
	  6 glxinfo:gdrv0            0.15     0.12
	  4 vulkani:disk$0           0.15     0.12
	  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
	  2 glxinfo                  0.08     0.05
	  2 glxinfo:cs0              0.08     0.05
	  2 glxinfo:disk$0           0.07     0.04
	  2 glxinfo:sh0              0.07     0.04
	  2 glxinfo:shlo0            0.07     0.04
	  6 php                      0.06     0.08
	  1 lspci                    0.01     0.02
	  3 rm                       0.00     0.11
	  1 ps                       0.00     0.01
	 66 sh                       0.00     0.00
	 13 gcc                      0.00     0.00
	  8 stat                     0.00     0.00
	  8 systemd-detect-          0.00     0.00
	  6 gsettings                0.00     0.00
	  6 sleep                    0.00     0.00
	  5 dconf worker             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 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 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
144 maximum processes

Process structure for computation

      88678) cockroach        cpu=1 start=4.80  finish=111.41
        88679) cockroach        cpu=9 start=4.80  finish=112.07
          88681) cockroach        cpu=1 start=4.80  finish=112.07
            88791) cockroach        cpu=8 start=60.34 finish=112.07
            88792) cockroach        cpu=0 start=62.27 finish=112.07
          88682) cockroach        cpu=10 start=4.80  finish=112.07
            88685) cockroach        cpu=5 start=4.80  finish=112.07
            88687) cockroach        cpu=5 start=4.81  finish=112.07
            88689) cockroach        cpu=3 start=4.81  finish=112.07
          88683) cockroach        cpu=0 start=4.80  finish=112.07
          88684) cockroach        cpu=12 start=4.80  finish=112.07
            88688) cockroach        cpu=1 start=4.81  finish=112.07
              88691) cockroach        cpu=6 start=4.81  finish=112.07
                88695) cockroach        cpu=11 start=4.83  finish=112.07
                  88696) cockroach        cpu=15 start=4.83  finish=112.07
                88697) cockroach        cpu=14 start=4.83  finish=112.07
                88787) cockroach        cpu=5 start=11.85 finish=112.07
              88788) cockroach        cpu=2 start=13.10 finish=112.07
              88789) cockroach        cpu=0 start=18.24 finish=112.07
          88686) cockroach        cpu=12 start=4.81  finish=112.07
          88690) cockroach        cpu=13 start=4.81  finish=112.07
            88692) cockroach        cpu=8 start=4.82  finish=112.07
              88693) cockroach        cpu=9 start=4.82  finish=112.07
                88694) cockroach        cpu=10 start=4.82  finish=112.07
              88698) cockroach        cpu=7 start=4.85  finish=112.07
                88699) cockroach        cpu=3 start=4.85  finish=112.07
                88790) cockroach        cpu=1 start=23.34 finish=112.07
          88700) cockroach        cpu=4 start=4.87  finish=112.07
        88680) sleep            cpu=13 start=4.80  finish=9.80 
        88701) cockroach        cpu=9 start=9.80  finish=110.36
          88702) cockroach        cpu=4 start=9.80  finish=110.36
            88722) cockroach        cpu=4 start=10.22 finish=110.36
            88723) cockroach        cpu=10 start=10.22 finish=110.36
            88724) cockroach        cpu=5 start=10.22 finish=110.36
            88725) cockroach        cpu=11 start=10.22 finish=110.36
            88726) cockroach        cpu=15 start=10.22 finish=110.36
            88727) cockroach        cpu=6 start=10.22 finish=110.36
            88728) cockroach        cpu=2 start=10.22 finish=110.36
            88729) cockroach        cpu=12 start=10.22 finish=110.36
            88730) cockroach        cpu=9 start=10.22 finish=110.36
            88731) cockroach        cpu=4 start=10.22 finish=110.36
            88732) cockroach        cpu=0 start=10.22 finish=110.36
            88733) cockroach        cpu=6 start=10.23 finish=110.36
            88734) cockroach        cpu=7 start=10.23 finish=110.36
            88735) cockroach        cpu=5 start=10.23 finish=110.36
            88736) cockroach        cpu=4 start=10.23 finish=110.36
            88737) cockroach        cpu=1 start=10.23 finish=110.35
            88738) cockroach        cpu=5 start=10.23 finish=110.35
            88739) cockroach        cpu=2 start=10.23 finish=110.35
            88740) cockroach        cpu=10 start=10.23 finish=110.35
            88741) cockroach        cpu=2 start=10.23 finish=110.35
            88742) cockroach        cpu=2 start=10.23 finish=110.35
            88743) cockroach        cpu=15 start=10.23 finish=110.35
            88744) cockroach        cpu=9 start=10.23 finish=110.35
            88745) cockroach        cpu=10 start=10.23 finish=110.35
            88746) cockroach        cpu=5 start=10.23 finish=110.35
            88747) cockroach        cpu=3 start=10.23 finish=110.35
            88748) cockroach        cpu=1 start=10.23 finish=110.35
            88749) cockroach        cpu=5 start=10.23 finish=110.35
            88750) cockroach        cpu=11 start=10.23 finish=110.35
            88751) cockroach        cpu=0 start=10.23 finish=110.35
            88752) cockroach        cpu=10 start=10.23 finish=110.35
            88753) cockroach        cpu=7 start=10.23 finish=110.35
            88754) cockroach        cpu=5 start=10.23 finish=110.35
            88755) cockroach        cpu=10 start=10.23 finish=110.35
            88756) cockroach        cpu=3 start=10.23 finish=110.35
            88757) cockroach        cpu=9 start=10.23 finish=110.35
            88758) cockroach        cpu=7 start=10.23 finish=110.35
            88759) cockroach        cpu=5 start=10.23 finish=110.35
            88760) cockroach        cpu=2 start=10.23 finish=110.35
            88761) cockroach        cpu=7 start=10.23 finish=110.35
            88762) cockroach        cpu=5 start=10.23 finish=110.35
            88763) cockroach        cpu=0 start=10.23 finish=110.35
            88764) cockroach        cpu=10 start=10.23 finish=110.35
            88765) cockroach        cpu=12 start=10.23 finish=110.35
            88766) cockroach        cpu=3 start=10.23 finish=110.35
            88767) cockroach        cpu=0 start=10.23 finish=110.35
            88768) cockroach        cpu=7 start=10.23 finish=110.35
            88769) cockroach        cpu=12 start=10.23 finish=110.35
            88770) cockroach        cpu=7 start=10.23 finish=110.35
            88771) cockroach        cpu=14 start=10.23 finish=110.35
            88772) cockroach        cpu=7 start=10.23 finish=110.35
            88773) cockroach        cpu=5 start=10.23 finish=110.35
            88774) cockroach        cpu=9 start=10.23 finish=110.35
            88775) cockroach        cpu=7 start=10.23 finish=110.35
            88776) cockroach        cpu=8 start=10.23 finish=110.35
            88777) cockroach        cpu=9 start=10.23 finish=110.35
            88778) cockroach        cpu=14 start=10.23 finish=110.35
            88779) cockroach        cpu=0 start=10.23 finish=110.35
            88780) cockroach        cpu=2 start=10.23 finish=110.35
            88781) cockroach        cpu=8 start=10.23 finish=110.35
            88782) cockroach        cpu=12 start=10.23 finish=110.35
            88783) cockroach        cpu=4 start=10.23 finish=110.35
            88784) cockroach        cpu=11 start=10.23 finish=110.35
          88703) cockroach        cpu=2 start=9.80  finish=110.36
            88705) cockroach        cpu=4 start=9.81  finish=110.36
            88708) cockroach        cpu=9 start=9.81  finish=110.36
              88714) cockroach        cpu=0 start=9.82  finish=110.36
            88710) cockroach        cpu=5 start=9.81  finish=110.36
              88715) cockroach        cpu=15 start=9.82  finish=110.36
                88716) cockroach        cpu=9 start=9.82  finish=110.36
              88717) cockroach        cpu=10 start=9.83  finish=110.36
                88718) cockroach        cpu=4 start=9.83  finish=110.36
                  88785) cockroach        cpu=1 start=10.29 finish=110.35
          88704) cockroach        cpu=14 start=9.81  finish=110.36
          88706) cockroach        cpu=11 start=9.81  finish=110.36
            88709) cockroach        cpu=8 start=9.81  finish=110.36
              88711) cockroach        cpu=0 start=9.81  finish=110.36
                88712) cockroach        cpu=5 start=9.81  finish=110.36
                  88719) cockroach        cpu=3 start=9.85  finish=110.36
                    88720) cockroach        cpu=7 start=9.85  finish=110.36
              88713) cockroach        cpu=7 start=9.81  finish=110.36
          88707) cockroach        cpu=8 start=9.81  finish=110.36
            88721) cockroach        cpu=12 start=9.87  finish=110.36
        88793) sleep            cpu=8 start=110.37 finish=111.37
        88794) rm               cpu=2 start=111.37 finish=111.41