Metrics (Intel) - phoronix/mafftThis test performs an alignment of 100 pyruvate decarboxylase sequences.
sh - pid 7701 On_CPU 0.858 On_Core 6.867 IPC 1.304 Retire 0.439 (43.9%) FrontEnd 0.175 (17.5%) Spec 0.308 (30.8%) Backend 0.078 (7.8%) Elapsed 4.37 Procs 145 Maxrss 172K Minflt 80085 Majflt 0 Inblock 0 Oublock 3784 Msgsnd 0 Msgrcv 0 Nsignals 0 Nvcsw 49796 (96.3%) Nivcsw 1909 Utime 29.281264 Stime 0.727976 Start 47787.64 Finish 47792.01
The program has a lot of little processes running and an overall On_CPU of 85%. There is a high level of speculative misses. Otherwise the IPC is medium and without much backend stalls. The process itself takes only 4 1/2 seconds. The high percentage of voluntary context switches also suggest I/O.
Metrics (AMD) - phoronix/mafftsh - pid 31571 On_CPU 0.513 On_Core 8.205 IPC 1.537 FrontCyc 0.042 (4.2%) BackCyc 0.380 (38.0%) Elapsed 3.81 Procs 161 Maxrss 168K Minflt 90940 Majflt 0 Inblock 0 Oublock 3784 Msgsnd 0 Msgrcv 0 Nsignals 0 Nvcsw 71232 (96.0%) Nivcsw 2987 Utime 30.643080 Stime 0.618453 Start 116989.42 Finish 116993.23
IPC on AMD is just slightly higher.
Process Tree - phoronix/mafftProcess Tree
7701) sh elapsed=4.37 start=0.00 finish=4.37
7702) mafft elapsed=4.37 start=0.00 finish=4.37
7703) mafft elapsed=4.37 start=0.00 finish=4.37
7704) dirname elapsed=0.00 start=0.01 finish=0.01
7705) basename elapsed=0.00 start=0.01 finish=0.01
7706) uname elapsed=0.00 start=0.01 finish=0.01
7707) basename elapsed=0.00 start=0.01 finish=0.01
7708) mafft elapsed=0.00 start=0.01 finish=0.01
7709) mafft elapsed=0.00 start=0.01 finish=0.01
7710) grep elapsed=0.00 start=0.01 finish=0.01
7711) mafft elapsed=0.00 start=0.01 finish=0.01
7712) mafft elapsed=0.00 start=0.01 finish=0.01
7713) grep elapsed=0.00 start=0.01 finish=0.01
7714) mafft elapsed=0.00 start=0.02 finish=0.02
7715) mafft elapsed=0.00 start=0.02 finish=0.02
7716) grep elapsed=0.00 start=0.02 finish=0.02
7717) mafft elapsed=0.00 start=0.02 finish=0.02
7718) mafft elapsed=0.00 start=0.02 finish=0.02
7719) grep elapsed=0.00 start=0.02 finish=0.02
7720) mafft elapsed=0.00 start=0.02 finish=0.02
7721) mafft elapsed=0.00 start=0.02 finish=0.02
7722) grep elapsed=0.00 start=0.02 finish=0.02
7723) mafft elapsed=0.00 start=0.02 finish=0.02
7724) mafft elapsed=0.00 start=0.02 finish=0.02
7725) grep elapsed=0.00 start=0.02 finish=0.02
7726) version elapsed=0.00 start=0.02 finish=0.02
7727) expr elapsed=0.00 start=0.02 finish=0.02
7728) expr elapsed=0.00 start=0.04 finish=0.04
7729) expr elapsed=0.00 start=0.04 finish=0.04
7730) expr elapsed=0.00 start=0.04 finish=0.04
7731) expr elapsed=0.00 start=0.04 finish=0.04
7732) mktemp elapsed=0.00 start=0.04 finish=0.04
7733) mafft elapsed=0.00 start=0.04 finish=0.04
7734) cat elapsed=0.00 start=0.04 finish=0.04
7735) tr elapsed=0.00 start=0.04 finish=0.04
7736) cat elapsed=0.00 start=0.04 finish=0.04
7737) tr elapsed=0.00 start=0.04 finish=0.04
7738) grep elapsed=0.00 start=0.04 finish=0.04
7739) cat elapsed=0.00 start=0.04 finish=0.04
7740) tr elapsed=0.00 start=0.04 finish=0.04
7741) grep elapsed=0.00 start=0.04 finish=0.04
7742) cat elapsed=0.00 start=0.04 finish=0.04
7743) tr elapsed=0.00 start=0.04 finish=0.04
7744) grep elapsed=0.01 start=0.04 finish=0.05
7745) cat elapsed=0.00 start=0.05 finish=0.05
7746) tr elapsed=0.00 start=0.05 finish=0.05
7747) grep elapsed=0.00 start=0.05 finish=0.05
7748) cat elapsed=0.00 start=0.05 finish=0.05
7749) tr elapsed=0.00 start=0.05 finish=0.05
7750) grep elapsed=0.00 start=0.05 finish=0.05
7751) cat elapsed=0.00 start=0.05 finish=0.05
7752) tr elapsed=0.00 start=0.05 finish=0.05
7753) grep elapsed=0.00 start=0.05 finish=0.05
7754) cat elapsed=0.00 start=0.05 finish=0.05
7755) tr elapsed=0.00 start=0.05 finish=0.05
7756) grep elapsed=0.00 start=0.05 finish=0.05
7757) cat elapsed=0.00 start=0.05 finish=0.05
7758) tr elapsed=0.00 start=0.05 finish=0.05
7759) grep elapsed=0.00 start=0.05 finish=0.05
7760) cat elapsed=0.00 start=0.05 finish=0.05
7761) tr elapsed=0.00 start=0.05 finish=0.05
7762) grep elapsed=0.00 start=0.05 finish=0.05
7763) file elapsed=0.01 start=0.05 finish=0.06
7764) grep elapsed=0.01 start=0.05 finish=0.06
7765) file elapsed=0.00 start=0.06 finish=0.06
7766) grep elapsed=0.00 start=0.06 finish=0.06
7767) file elapsed=0.00 start=0.06 finish=0.06
7768) grep elapsed=0.00 start=0.06 finish=0.06
7769) file elapsed=0.00 start=0.06 finish=0.06
7770) grep elapsed=0.00 start=0.06 finish=0.06
7771) file elapsed=0.00 start=0.06 finish=0.06
7772) grep elapsed=0.00 start=0.06 finish=0.06
7773) file elapsed=0.00 start=0.06 finish=0.06
7774) grep elapsed=0.00 start=0.06 finish=0.06
7775) file elapsed=0.01 start=0.06 finish=0.07
7776) grep elapsed=0.01 start=0.06 finish=0.07
7777) file elapsed=0.00 start=0.07 finish=0.07
7778) grep elapsed=0.00 start=0.07 finish=0.07
7779) file elapsed=0.00 start=0.07 finish=0.07
7780) grep elapsed=0.00 start=0.07 finish=0.07
7781) file elapsed=0.00 start=0.07 finish=0.07
7782) grep elapsed=0.00 start=0.07 finish=0.07
7783) file elapsed=0.00 start=0.07 finish=0.07
7784) grep elapsed=0.00 start=0.07 finish=0.07
7785) file elapsed=0.00 start=0.07 finish=0.07
7786) grep elapsed=0.00 start=0.07 finish=0.07
7787) file elapsed=0.00 start=0.07 finish=0.07
7788) grep elapsed=0.00 start=0.07 finish=0.07
7789) file elapsed=0.00 start=0.07 finish=0.07
7790) grep elapsed=0.00 start=0.07 finish=0.07
7791) file elapsed=0.00 start=0.07 finish=0.07
7792) grep elapsed=0.00 start=0.07 finish=0.07
7793) file elapsed=0.00 start=0.07 finish=0.07
7794) grep elapsed=0.00 start=0.07 finish=0.07
7795) file elapsed=0.00 start=0.07 finish=0.07
7796) grep elapsed=0.00 start=0.07 finish=0.07
7797) awk elapsed=0.00 start=0.07 finish=0.07
7798) awk elapsed=0.00 start=0.08 finish=0.08
7799) awk elapsed=0.00 start=0.08 finish=0.08
7800) awk elapsed=0.00 start=0.08 finish=0.08
7801) awk elapsed=0.00 start=0.08 finish=0.08
7802) awk elapsed=0.00 start=0.08 finish=0.08
7803) awk elapsed=0.00 start=0.08 finish=0.08
7804) awk elapsed=0.00 start=0.08 finish=0.08
7805) awk elapsed=0.00 start=0.08 finish=0.08
7806) awk elapsed=0.00 start=0.08 finish=0.08
7807) awk elapsed=0.00 start=0.08 finish=0.08
7808) mafft elapsed=0.00 start=0.08 finish=0.08
7809) grep elapsed=0.00 start=0.08 finish=0.08
7810) head elapsed=0.00 start=0.08 finish=0.08
7811) mafft elapsed=0.00 start=0.08 finish=0.08
7812) grep elapsed=0.00 start=0.08 finish=0.08
7813) head elapsed=0.00 start=0.08 finish=0.08
7814) awk elapsed=0.00 start=0.09 finish=0.09
7815) cat elapsed=0.00 start=0.09 finish=0.09
7816) cat elapsed=0.00 start=0.09 finish=0.09
7817) tbfast elapsed=1.87 start=0.09 finish=1.96
7818) tbfast elapsed=1.69 start=0.09 finish=1.78
7819) tbfast elapsed=1.69 start=0.09 finish=1.78
7820) tbfast elapsed=1.69 start=0.09 finish=1.78
7821) tbfast elapsed=1.69 start=0.09 finish=1.78
7822) tbfast elapsed=1.69 start=0.09 finish=1.78
7823) tbfast elapsed=1.69 start=0.09 finish=1.78
7824) tbfast elapsed=1.69 start=0.09 finish=1.78
7825) tbfast elapsed=1.69 start=0.09 finish=1.78
7826) tbfast elapsed=0.09 start=1.81 finish=1.90
7827) tbfast elapsed=0.09 start=1.81 finish=1.90
7828) tbfast elapsed=0.09 start=1.81 finish=1.90
7829) tbfast elapsed=0.09 start=1.81 finish=1.90
7830) tbfast elapsed=0.10 start=1.81 finish=1.91
7831) tbfast elapsed=0.11 start=1.81 finish=1.92
7832) tbfast elapsed=0.14 start=1.81 finish=1.95
7833) tbfast elapsed=0.11 start=1.81 finish=1.92
7834) dvtditr elapsed=2.41 start=1.96 finish=4.37
7835) dvtditr elapsed=2.36 start=2.00 finish=4.36
7836) dvtditr elapsed=2.37 start=2.00 finish=4.37
7837) dvtditr elapsed=2.37 start=2.00 finish=4.37
7838) dvtditr elapsed=2.37 start=2.00 finish=4.37
7839) dvtditr elapsed=2.37 start=2.00 finish=4.37
7840) dvtditr elapsed=2.37 start=2.00 finish=4.37
7841) dvtditr elapsed=2.37 start=2.00 finish=4.37
7842) dvtditr elapsed=2.37 start=2.00 finish=4.37
7843) dvtditr elapsed=2.37 start=2.00 finish=4.37
7844) cat elapsed=0.00 start=4.37 finish=4.37
7845) rm elapsed=0.00 start=4.37 finish=4.37
The bulk of the time spent is running parallel instances of tbfast for 1.87 seconds and dvtditr for 2.37 seconds. Otherwise many small processes run and also some idle times.
Adding time from all processes shows gaps that account for only 85% On_CPU.
About this graph
Individual cores look same as total.
The IPC is moderately high.
About this graph
Speculative misses are the largest issue with lower levels of backend stalls.
retire 0.523 ms_uops 0.005 speculation 0.183 branch_misses 88.08% machine_clears 11.92% frontend 0.169 idq_uops_delivered_0 0.026 icache_stall 0.003 itlb_misses 0.000 idq_uops_delivered_1 0.052 idq_uops_delivered_2 0.112 idq_uops_delivered_3 0.146 dsb_ops 84.70% backend 0.124 resource_stalls.sb 0.029 stalls_ldm_pending 0.216
Speculative misses are mostly branch misses. The numbers are slightly different because it also includes the phoronix test suite actions and the benchmark is quick running.
Next steps: Understand branch misses.