{"id":2004,"date":"2024-03-05T12:15:23","date_gmt":"2024-03-05T12:15:23","guid":{"rendered":"https:\/\/mvermeulen.org\/perf\/?page_id=2004"},"modified":"2024-03-05T13:00:57","modified_gmt":"2024-03-05T13:00:57","slug":"ebizzy","status":"publish","type":"page","link":"https:\/\/mvermeulen.org\/perf\/workloads\/phoronix\/ebizzy\/","title":{"rendered":"ebizzy"},"content":{"rendered":"\n

A test that generates workloads resembling web server workloads. Quick running but many threads.<\/p>\n\n\n\n

\"\"<\/figure>\n\n\n\n

Topdown profile shows higher levels of both backend stalls and frontend stalls.<\/p>\n\n\n\n

\"\"<\/figure>\n\n\n\n

AMD metrics show a large amount of L2 access? Not much floating point and a high opcache miss rate.<\/p>\n\n\n\n

elapsed              76.014\non_cpu               0.790          # 12.64 \/ 16 cores\nutime                960.128\nstime                0.969\nnvcsw                2441           # 3.57%\nnivcsw               65967          # 96.43%\ninblock              0              # 0.00\/sec\nonblock              12568          # 165.34\/sec\ncpu-clock            961212794580   # 961.213 seconds\ntask-clock           961219693426   # 961.220 seconds\npage faults          174119         # 181.144\/sec\ncontext switches     68613          # 71.381\/sec\ncpu migrations       794            # 0.826\/sec\nmajor page faults    43             # 0.045\/sec\nminor page faults    174076         # 181.099\/sec\nalignment faults     0              # 0.000\/sec\nemulation faults     0              # 0.000\/sec\nbranches             8490158192     # 233.934 branches per 1000 inst\nbranch misses        122343044      # 1.44% branch miss\nconditional          6245003590     # 172.072 conditional branches per 1000 inst\nindirect             174774988      # 4.816 indirect branches per 1000 inst\ncpu-cycles           3619643263703  # 2.97 GHz\ninstructions         36257254806    # 0.01 IPC low\nslots                7241048093232  #\nretiring             481361243242   #  6.6% ( 7.1%) low\n-- ucode             56944409       #     0.0%\n-- fastpath          481304298833   #     6.6%\nfrontend             2366660898335  # 32.7% (35.1%)\n-- latency           70688934948    #     1.0%\n-- bandwidth         2295971963387  #    31.7%\nbackend              3883706267866  # 53.6% (57.7%)\n-- cpu               149362501021   #     2.1%\n-- memory            3734343766845  #    51.6%\nspeculation          2804700251     #  0.0% ( 0.0%) low\n-- branch mispredict 2049764610     #     0.0%\n-- pipeline restart  754935641      #     0.0%\nsmt-contention       506509742595   #  7.0% ( 0.0%)\ncpu-cycles           3614193928091  # 2.96 GHz\ninstructions         36147971263    # 0.01 IPC low\ninstructions         11965344127    # 42861.333 l2 access per 1000 inst\nl2 hit from l1       228293533402   # 26.95% l2 miss\nl2 miss from l1      20386259467    #\nl2 hit from l2 pf    166706468197   #\nl3 hit from l2 pf    117806766518   #\nl3 miss from l2 pf   43833622       #\ninstructions         12235477579    # 18.695 float per 1000 inst\nfloat 512            66             # 0.000 AVX-512 per 1000 inst\nfloat 256            596            # 0.000 AVX-256 per 1000 inst\nfloat 128            228741691      # 18.695 AVX-128 per 1000 inst\nfloat MMX            0              # 0.000 MMX per 1000 inst\nfloat scalar         0              # 0.000 scalar per 1000 inst\ninstructions         36020965289    #\nopcache              9790344647     # 271.796 opcache per 1000 inst\nopcache miss         4930331972     # 50.4% opcache miss rate\nl1 dTLB miss         11689841338    # 324.529 L1 dTLB per 1000 inst\nl2 dTLB miss         26979105       # 0.749 L2 dTLB per 1000 inst\ninstructions         36020505704    #\nicache               6350080470     # 176.291 icache per 1000 inst\nicache miss          2161166540     # 34.0% icache miss rate\nl1 iTLB miss         9196328        # 0.255 L1 iTLB per 1000 inst\nl2 iTLB miss         0              # 0.000 L2 iTLB per 1000 inst\ntlb flush            19035          # 0.001 TLB flush per 1000 inst\n<\/code><\/pre>\n\n\n\n
Intel metrics also show a high amount of L2 access and a very large amount of backend memory stalls.<\/p>\n\n\n\n
elapsed              76.197\non_cpu               0.788          # 12.61 \/ 16 cores\nutime                960.003\nstime                0.625\nnvcsw                2219           # 3.37%\nnivcsw               63631          # 96.63%\ninblock              2216           # 29.08\/sec\nonblock              1320           # 17.32\/sec\ncpu-clock            960779969649   # 960.780 seconds\ntask-clock           960785636136   # 960.786 seconds\npage faults          162859         # 169.506\/sec\ncontext switches     66055          # 68.751\/sec\ncpu migrations       855            # 0.890\/sec\nmajor page faults    24             # 0.025\/sec\nminor page faults    162835         # 169.481\/sec\nalignment faults     0              # 0.000\/sec\nemulation faults     0              # 0.000\/sec\nbranches             5326251888     # 224.184 branches per 1000 inst\nbranch misses        21464693       # 0.40% branch miss\nconditional          5326266800     # 224.185 conditional branches per 1000 inst\nindirect             656377561      # 27.627 indirect branches per 1000 inst\nslots                5160889464836  #\nretiring             291731107053   #  5.7% ( 5.7%) low\n-- ucode             100764509180   #     2.0%\n-- fastpath          190966597873   #     3.7%\nfrontend             346184570707   #  6.7% ( 6.7%)\n-- latency           337244658529   #     6.5%\n-- bandwidth         8939912178     #     0.2%\nbackend              4521787284195  # 87.6% (87.6%) high\n-- cpu               572017397254   #    11.1%\n-- memory            3949769886941  #    76.5%\nspeculation          3292312798     #  0.1% ( 0.1%) low\n-- branch mispredict 3103520615     #     0.1%\n-- pipeline restart  188792183      #     0.0%\nsmt-contention       0              #  0.0% ( 0.0%)\ncpu-cycles           3597502784460  # 1.82 GHz\ninstructions         32690767154    # 0.01 IPC low\nl2 access            547037131507   # 21244.104 l2 access per 1000 inst\nl2 miss              212437512662   # 38.83% l2 miss\ncpu-cycles           5628197624978  # 85.4% memory latency\nload stalls          4804943018686  # 24.1% l1 bound\nl1 miss              3448039312113  # 38.0% l2 bound\nl2 miss              1310663434850  # 23.3% l3 bound\nl3 miss              956314300      #  0.0% dram bound\nstore_stalls         3981547654     #  0.1% store bound\n<\/code><\/pre>\n\n\n\n
Process summary shows time in the ebizzy application.<\/p>\n\n\n\n
824 processes\n\t476 ebizzy               147756.72    17.82\n\t 68 clinfo                  14.88     8.31\n\t 38 vulkaninfo               1.33     1.14\n\t  4 vulkani:disk$0           0.15     0.12\n\t  6 php                      0.09     0.14\n\t  2 llvmpipe-0               0.08     0.06\n\t  2 llvmpipe-1               0.08     0.06\n\t  2 llvmpipe-10              0.08     0.06\n\t  2 llvmpipe-11              0.08     0.06\n\t  2 llvmpipe-12              0.08     0.06\n\t  2 llvmpipe-13              0.08     0.06\n\t  2 llvmpipe-14              0.08     0.06\n\t  2 llvmpipe-15              0.08     0.06\n\t  2 llvmpipe-2               0.08     0.06\n\t  2 llvmpipe-3               0.08     0.06\n\t  2 llvmpipe-4               0.08     0.06\n\t  2 llvmpipe-5               0.08     0.06\n\t  2 llvmpipe-6               0.08     0.06\n\t  2 llvmpipe-7               0.08     0.06\n\t  2 llvmpipe-8               0.08     0.06\n\t  2 llvmpipe-9               0.08     0.06\n\t  6 glxinfo:gdrv0            0.07     0.11\n\t  6 glxinfo:gl0              0.07     0.11\n\t  6 clang                    0.06     0.06\n\t  2 glxinfo                  0.05     0.04\n\t  2 glxinfo:cs0              0.05     0.04\n\t  2 glxinfo:disk$0           0.05     0.04\n\t  2 glxinfo:sh0              0.05     0.04\n\t  2 glxinfo:shlo0            0.05     0.04\n\t  1 lspci                    0.00     0.02\n\t 82 sh                       0.00     0.00\n\t 13 gcc                      0.00     0.00\n\t 13 gsettings                0.00     0.00\n\t  8 stat                     0.00     0.00\n\t  8 systemd-detect-          0.00     0.00\n\t  6 llvm-link                0.00     0.00\n\t  5 phoronix-test-s          0.00     0.00\n\t  3 rocminfo                 0.00     0.00\n\t  2 cc                       0.00     0.00\n\t  2 gmain                    0.00     0.00\n\t  2 lscpu                    0.00     0.00\n\t  2 uname                    0.00     0.00\n\t  2 which                    0.00     0.00\n\t  2 xset                     0.00     0.00\n\t  1 date                     0.00     0.00\n\t  1 dconf worker             0.00     0.00\n\t  1 dirname                  0.00     0.00\n\t  1 dmesg                    0.00     0.00\n\t  1 dmidecode                0.00     0.00\n\t  1 grep                     0.00     0.00\n\t  1 ifconfig                 0.00     0.00\n\t  1 ip                       0.00     0.00\n\t  1 lsmod                    0.00     0.00\n\t  1 mktemp                   0.00     0.00\n\t  1 ps                       0.00     0.00\n\t  1 qdbus                    0.00     0.00\n\t  1 readlink                 0.00     0.00\n\t  1 realpath                 0.00     0.00\n\t  1 sed                      0.00     0.00\n\t  1 sort                     0.00     0.00\n\t  1 stty                     0.00     0.00\n\t  1 systemctl                0.00     0.00\n\t  1 template.sh              0.00     0.00\n\t  1 wc                       0.00     0.00\n\t  1 xrandr                   0.00     0.00\n0 processes running\n47 maximum processes\n<\/code><\/pre>\n\n\n\n
Computation blocks look as follows<\/p>\n\n\n\n
      958113) ebizzy           cpu=6 start=5.64  finish=25.65\n        958114) ebizzy           cpu=15 start=5.64  finish=25.65\n          958115) ebizzy           cpu=9 start=5.64  finish=25.65\n          958116) ebizzy           cpu=12 start=5.64  finish=25.65\n          958117) ebizzy           cpu=2 start=5.64  finish=25.65\n          958118) ebizzy           cpu=0 start=5.64  finish=25.65\n          958119) ebizzy           cpu=7 start=5.64  finish=25.65\n          958120) ebizzy           cpu=0 start=5.64  finish=25.65\n          958121) ebizzy           cpu=4 start=5.64  finish=25.65\n          958122) ebizzy           cpu=2 start=5.64  finish=25.64\n          958123) ebizzy           cpu=4 start=5.64  finish=25.64\n          958124) ebizzy           cpu=10 start=5.64  finish=25.64\n          958125) ebizzy           cpu=12 start=5.64  finish=25.64\n          958126) ebizzy           cpu=13 start=5.64  finish=25.64\n          958127) ebizzy           cpu=8 start=5.64  finish=25.64\n          958128) ebizzy           cpu=11 start=5.64  finish=25.64\n          958129) ebizzy           cpu=13 start=5.64  finish=25.64\n          958130) ebizzy           cpu=1 start=5.64  finish=25.64\n          958131) ebizzy           cpu=4 start=5.64  finish=25.64\n          958132) ebizzy           cpu=15 start=5.64  finish=25.64\n          958133) ebizzy           cpu=3 start=5.64  finish=25.64\n          958134) ebizzy           cpu=14 start=5.64  finish=25.64\n          958135) ebizzy           cpu=3 start=5.64  finish=25.64\n          958136) ebizzy           cpu=10 start=5.64  finish=25.64\n          958137) ebizzy           cpu=11 start=5.64  finish=25.64\n          958138) ebizzy           cpu=6 start=5.64  finish=25.64\n          958139) ebizzy           cpu=1 start=5.64  finish=25.64\n          958140) ebizzy           cpu=6 start=5.64  finish=25.64\n          958141) ebizzy           cpu=5 start=5.64  finish=25.64\n          958142) ebizzy           cpu=15 start=5.64  finish=25.64\n          958143) ebizzy           cpu=9 start=5.64  finish=25.64\n          958144) ebizzy           cpu=14 start=5.64  finish=25.64\n          958145) ebizzy           cpu=8 start=5.64  finish=25.64\n          958146) ebizzy           cpu=7 start=5.64  finish=25.64\n<\/code><\/pre>\n","protected":false},"excerpt":{"rendered":"
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 …<\/span> Continue reading →<\/span><\/a><\/p>\n","protected":false},"author":1,"featured_media":0,"parent":58,"menu_order":0,"comment_status":"closed","ping_status":"closed","template":"","meta":{"footnotes":""},"class_list":["post-2004","page","type-page","status-publish","hentry"],"_links":{"self":[{"href":"https:\/\/mvermeulen.org\/perf\/wp-json\/wp\/v2\/pages\/2004","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/mvermeulen.org\/perf\/wp-json\/wp\/v2\/pages"}],"about":[{"href":"https:\/\/mvermeulen.org\/perf\/wp-json\/wp\/v2\/types\/page"}],"author":[{"embeddable":true,"href":"https:\/\/mvermeulen.org\/perf\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/mvermeulen.org\/perf\/wp-json\/wp\/v2\/comments?post=2004"}],"version-history":[{"count":2,"href":"https:\/\/mvermeulen.org\/perf\/wp-json\/wp\/v2\/pages\/2004\/revisions"}],"predecessor-version":[{"id":2013,"href":"https:\/\/mvermeulen.org\/perf\/wp-json\/wp\/v2\/pages\/2004\/revisions\/2013"}],"up":[{"embeddable":true,"href":"https:\/\/mvermeulen.org\/perf\/wp-json\/wp\/v2\/pages\/58"}],"wp:attachment":[{"href":"https:\/\/mvermeulen.org\/perf\/wp-json\/wp\/v2\/media?parent=2004"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}