OV - Compare Index

Help is available by moving the cursor above any symbol or by checking MAQAO website.

▶Compared Reports

r0: G3 ACfL best (O3)
r1: G4 ACfL best (O3)

Global Metrics

Metric		r0	r1
Total Time (s)		12.7 E3	7.91 E3
Max (Thread Active Time) (s)		12.7 E3	7.88 E3
Average Active Time (s)		12.4 E3	7.77 E3
Activity Ratio (%)		97.6	98.1
Average number of active threads		62.471	94.193
Affinity Stability (%)		100.0	100.0
Time in analyzed loops (%)		79.1	78.0
Time in analyzed innermost loops (%)		69.6	71.4
Time in user code (%)		83.3	80.8
Compilation Options Score (%)		99.5	99.5
Array Access Efficiency (%)		47.7	65.1

Potential Speedups
Perfect Flow Complexity		1.02	1.02
Perfect OpenMP + MPI + Pthread		1.09	1.09
Perfect OpenMP + MPI + Pthread + Perfect Load Distribution		1.18	1.21
No Scalar Integer	Potential Speedup	1.18	1.17
No Scalar Integer	Nb Loops to get 80%	24	25
FP Vectorised	Potential Speedup	1.08	1.04
FP Vectorised	Nb Loops to get 80%	16	11
Fully Vectorised	Potential Speedup	1.50	1.20
Fully Vectorised	Nb Loops to get 80%	38	39
Only FP Arithmetic	Potential Speedup	1.53	1.57
Only FP Arithmetic	Nb Loops to get 80%	41	41

Application Categorization

Time

0.00

2000.00

4000.00

6000.00

8000.00

10000.00

12000.00

14000.00

System

Binary

MPI

Math

OMP

String

Pthread

Memory

Others

0.00

2000.00

4000.00

6000.00

8000.00

10000.00

12000.00

14000.00

System

Binary

MPI

Math

OMP

String

Pthread

Memory

Others

Coverage

0.00

16.67

33.33

50.00

66.67

83.33

100.00

System

Binary

MPI

Math

OMP

String

Pthread

Memory

Others

0.00

16.67

33.33

50.00

66.67

83.33

100.00

System

Binary

MPI

Math

OMP

String

Pthread

Memory

Others

Source Object	Issue
▼engine_linuxa64_ompi–
▼m2cplr.F–
○
▼c3deri3.F–
○
▼inter_count_node_curv.F–
○
▼accele.F–
○
▼cmain3.F–
○
▼sigeps27c.F–
○
▼c3fcum3.F–
○
▼s4deri3.F–
○
▼sigeps70.F–
○
▼cortdir3.F–
○
▼i11optcd.F–
○
▼i7dst3.F–
○
▼depla.F–
○
▼i7trc.F–
○
▼vitesse.F–
○
▼mqviscb.F–
○
▼sortie_main.F–
○
▼cbilan.F–
○
▼cdkcoor3.F–
○
▼c3dt3.F–
○
▼c3fint3.F–
○
▼czcorc.F–
○
▼get_volume_area.F90–
○
▼c3forc3.F–
○
▼spmd_cell_size_exchange.F–
○
▼sigeps36c.F–
○
▼spmd_exch2_a_pon.F–
○
▼dtnoda.F–
○
▼srota3.F–
○
▼intfop2.F–
○
▼i7mainf.F–
○
▼s4bilan.F–
○
▼czproj.F–
○
▼sigeps19c.F–
○
▼sfint43.F–
○
▼c3coef3.F–
○
▼spmd_exchi_a_pon.F–
○
▼c3stra3.F–
○
▼vinter.F–
○
▼sigeps02c.F–
○
▼i7ass3.F–
○
▼s4forc3.F–
○
▼czforc3.F–
○
▼spmd_i7xvcom2.F–
○
▼czfintce.F–
○
▼shvis3.F–
○
▼czcorp5.F–
○
▼c3coor3.F–
○
▼srho3.F–
○
▼ass2sort.F–
○
▼i2forces.F–
○
▼c3defo3.F–
○
▼c3evec3.F–
○
▼uroto.F–
○
▼srcoor3.F–
○
▼rgbodv.F–
○
▼i11cdcor3.F–
○
▼scoor43.F–
○
▼m27plas.F–
○
▼mmain.F90–
○
▼asspar4.F–
○
▼m27elas.F–
○
▼i11buce_crit.F–
○
▼damping.F–
○
▼inter_cell_color.F–
○
▼mulawglc.F–
○
▼i11dst3.F–
○
▼forintc.F–
○
▼c3curv3.F–
○
▼gravit.F–
○
▼i11ass3.F–
○
▼i7pen3.F–
○
▼i2vit3.F–
○
▼roto.F–
○
▼intti1.F–
○
▼mulaw.F90–
○
▼mulawc.F–
○
▼rgbcor.F–
○
▼i7for3.F–
○
▼i7trivox.F–
○
▼rgbodfp.F–
○
▼c3updt3.F–
○
▼scumu3p.F–
○
▼s4defo3.F–
○
▼scoor3.F–
○
▼fail_connect.F–
○
▼czfintn.F–
○
▼cncoef3.F–
○
▼i7sto.F–
○
▼czstra3.F–
○
▼bcs10.F–
○
▼i7cdcor3.F–
○
▼parit.F–
○
▼czdef.F–
○
▼deplafakeige.F–
○
▼i2for3p.F–
○
▼ecrit.F–
○
▼s4fint3.F–
○
▼sensor_energy_bilan.F–
○
▼inter_voxel_creation.F–
○
▼suser43.F–
○
▼i7optcd.F–
○
▼s4cumu3p.F–
○
▼i11for3.F–
○
▼resol.F–
○
▼i7cor3t.F–
○
▼suforc3.F–
○
▼stl_algo.h–
○	-g is missing for some functions (possibly ones added by the compiler), it is needed to have more accurate reports. Other recommended flags are: -O2/-O3, -march=(target)
▼cndt3.F–
○
▼i11mainf.F–
○
▼m27crak.F–
○
▼sigeps01g.F–
○
▼s4coor3.F–
○
▼table_mat_vinterp.F–
○
▼i7cor3.F–
○
▼cupdtn3.F–
○
▼sigeps59.F–
○
▼i7buce_crit.F–
○
▼i11cor3.F–
○
▼c3bilan.F–
○	-O2, -O3 or -Ofast is missing.
▼volpresp.F–
○
▼layini.F–
○

Source Object	Issue
▼engine_linuxa64_ompi–
▼m2cplr.F–
○
▼c3deri3.F–
○
▼inter_count_node_curv.F–
○
▼accele.F–
○
▼cmain3.F–
○
▼sigeps27c.F–
○
▼c3fcum3.F–
○
▼s4deri3.F–
○
▼mulawc.F–
○
▼sigeps70.F–
○
▼cbilan.F–
○
▼i7sto.F–
○
▼i11optcd.F–
○
▼i7dst3.F–
○
▼depla.F–
○
▼i7trc.F–
○
▼i7cor3.F–
○
▼mqviscb.F–
○
▼sortie_main.F–
○
▼dtnoda.F–
○
▼intti2.F–
○
▼resol.F–
○
▼i11dst3.F–
○
▼gravit.F–
○
▼get_volume_area.F90–
○
▼rgbodfp.F–
○
▼spmd_cell_size_exchange.F–
○
▼sigeps36c.F–
○
▼spmd_exch2_a_pon.F–
○
▼cncoef3.F–
○
▼srota3.F–
○
▼intfop2.F–
○
▼i7mainf.F–
○
▼i2forces.F–
○
▼czproj.F–
○
▼czcorc.F–
○
▼sfint43.F–
○
▼c3coef3.F–
○
▼i7pen3.F–
○
▼c3stra3.F–
○
▼vinter.F–
○
▼sigeps02c.F–
○
▼i7ass3.F–
○
▼s4forc3.F–
○
▼shvis3.F–
○
▼spmd_i7xvcom2.F–
○
▼czfintce.F–
○
▼sensor_energy_bilan.F–
○
▼czcorp5.F–
○
▼scumu3p.F–
○
▼srho3.F–
○
▼vitesse.F–
○
▼i11ass3.F–
○
▼c3defo3.F–
○
▼rgbodv.F–
○
▼uroto.F–
○
▼srcoor3.F–
○
▼roto.F–
○
▼i11cdcor3.F–
○
▼ecrit.F–
○
▼m27plas.F–
○
▼mmain.F90–
○
▼scoor43.F–
○
▼m27elas.F–
○
▼szhour3.F–
○
▼spmd_i7tool.F–
○
▼mulawglc.F–
○
▼c3fint3.F–
○
▼asspar4.F–
○
▼forintc.F–
○
▼ass2sort.F–
○
▼cdkcoor3.F–
○
▼sigeps19c.F–
○
▼inter_cell_color.F–
○
▼i2vit3.F–
○
▼czforc3.F–
○
▼intti1.F–
○
▼mulaw.F90–
○
▼bcs10.F–
○
▼rgbcor.F–
○
▼i7for3.F–
○
▼i7trivox.F–
○
▼m27crak.F–
○
▼c3updt3.F–
○
▼i11mainf.F–
○
▼s4defo3.F–
○
▼fail_connect.F–
○
▼c3forc3.F–
○
▼czfintn.F–
○
▼czdef.F–
○
▼i11buce_crit.F–
○
▼czstra3.F–
○
▼spmd_exchi_a_pon.F–
○
▼i7cdcor3.F–
○
▼parit.F–
○
▼layini.F–
○
▼deplafakeige.F–
○
▼i2for3p.F–
○
▼cndt3.F–
○
▼s4fint3.F–
○
▼suser43.F–
○
▼inter_voxel_creation.F–
○
▼damping.F–
○
▼i7optcd.F–
○
▼s4cumu3p.F–
○
▼i11for3.F–
○
▼scoor3.F–
○
▼i7cor3t.F–
○
▼suforc3.F–
○
▼stl_algo.h–
○	-g is missing for some functions (possibly ones added by the compiler), it is needed to have more accurate reports. Other recommended flags are: -O2/-O3, -march=(target)
▼c3dt3.F–
○
▼c3coor3.F–
○
▼c3curv3.F–
○
▼sigeps01g.F–
○
▼s4coor3.F–
○
▼table_mat_vinterp.F–
○
▼c3evec3.F–
○
▼cupdtn3.F–
○
▼s4bilan.F–
○
▼sigeps59.F–
○
▼i11cor3.F–
○
▼c3bilan.F–
○	-O2, -O3 or -Ofast is missing.
▼volpresp.F–
○
▼i7buce_crit.F–
○

Experiment Summaries

	r0	r1
Experiment Name
Application	/home/eoseret/Applications/OpenRadioss/exec/engine_linuxa64_ompi	/home/hbollore/pop3/openradioss/OpenRadioss/exec/engine_linuxa64_ompi
Timestamp	2024-12-17 16:56:15	2024-12-11 17:19:20
Experiment Type	MPI; OpenMP;	same as r0
Machine	ip-172-31-18-66	ip-172-31-47-249.ec2.internal
Architecture	aarch64	same as r0
Micro Architecture	ARM_NEOVERSE_V1	ARM_NEOVERSE_V2
Model Name
Cache Size
Number of Cores
Maximal Frequency	0 GHz	same as r0
OS Version	Linux 5.15.0-1072-aws #78~20.04.1-Ubuntu SMP Wed Oct 9 15:29:54 UTC 2024	Linux 6.1.109-118.189.amzn2023.aarch64 #1 SMP Tue Sep 10 08:58:40 UTC 2024
Architecture used during static analysis	aarch64	same as r0
Micro Architecture used during static analysis	ARM_NEOVERSE_V1	ARM_NEOVERSE_V2
Compilation Options	engine_linuxa64_ompi: Arm F90 F90 Flang - 1.5 2017-05-01 flang -I /home/eoseret/Applications/OpenRadioss/engine/../common_source/includes -I /home/eoseret/Applications/OpenRadioss/engine/../common_source/modules -I /home/eoseret/Applications/OpenRadioss/engine/share/includes -I /home/eoseret/Applications/OpenRadioss/engine/share/r8 -I /home/eoseret/Applications/OpenRadioss/engine/share/spe_inc -I /home/eoseret/Applications/OpenRadioss/engine/cbuild_engine_linuxa64_ompi/CMakeFiles/includes_engine_linuxa64_ompi -g -mcpu=native -fno-omit-frame-pointer -module CMakeFiles/modules_engine_linuxa64_ompi -D WITHOUT_LINALG -mcpu=native -D COMP_ARMFLANG=1 -D ARCH_CPU=ARM -fopenmp -D MYREAL8 -ffixed-line-length-none -D MPI -I /home/eoseret/openmpi_acfl2410/include -D CPP_mach=CPP_p4linux964 -D CPP_rel=70 -O3 -nofma -ffp-contract=off -fno-unsafe-math-optimizations -fno-fast-math -fveclib=none -I /home/eoseret/Applications/OpenRadioss/engine/../extlib/h3d/includes -c -o CMakeFiles/engine_linuxa64_ompi.dir/source/interfaces/intsort/i7optcd.F.o	engine_linuxa64_ompi: Arm F90 F90 Flang - 1.5 2017-05-01 flang -I /home/hbollore/pop3/openradioss/OpenRadioss/engine/../common_source/includes -I /home/hbollore/pop3/openradioss/OpenRadioss/engine/../common_source/modules -I /home/hbollore/pop3/openradioss/OpenRadioss/engine/share/includes -I /home/hbollore/pop3/openradioss/OpenRadioss/engine/share/r8 -I /home/hbollore/pop3/openradioss/OpenRadioss/engine/share/spe_inc -I /home/hbollore/pop3/openradioss/OpenRadioss/engine/cbuild_engine_linuxa64_ompi/CMakeFiles/includes_engine_linuxa64_ompi -g -mcpu=native -fno-omit-frame-pointer -module CMakeFiles/modules_engine_linuxa64_ompi -D WITHOUT_LINALG -mcpu=native -D COMP_ARMFLANG=1 -D ARCH_CPU=ARM -fopenmp -D MYREAL8 -ffixed-line-length-none -D MPI -I /home/hbollore/soft/openmpi-5.0.6-armsuite/include/ -D CPP_mach=CPP_p4linux964 -D CPP_rel=70 -O3 -nofma -ffp-contract=off -fno-unsafe-math-optimizations -fno-fast-math -fveclib=none -I /home/hbollore/pop3/openradioss/OpenRadioss/engine/../extlib/h3d/includes -c -o CMakeFiles/engine_linuxa64_ompi.dir/source/materials/mat_share/mulawc.F.o
Number of processes observed	32	48
Number of threads observed	64	96
Frequency Driver	NA	same as r0
Frequency Governor	NA	same as r0
Huge Pages	madvise	same as r0
Hyperthreading	off	same as r0
Number of sockets	1	same as r0
Number of cores per socket	64	96
MAQAO version	2.20.13	same as r0
MAQAO build	53f11931028ddbb6e7cc31119f63e8b014122ecb::20241210-180634	same as r0
Comments	OpenRadioss, dataset (Taurus10M short), compiled with ACfL 24.10 and running on AWS G3 with 32 MPI ranks and OMP_NUM_THREADS=2

Report Configuration

▶Compared Reports

Global Metrics

Cumulated Speedup If No Scalar Integer

Cumulated Speedup If FP Vectorized

Cumulated Speedup If Fully Vectorized

Cumulated Speedup If Only FP Arithmetic

Loop Based Profiles

Innermost / Single Loops

Inbetween Loops

Outermost Loops

Cumulated Coverage With All Loops

Innermost Loop Based Profiles

Coverage

Count

Application Categorization

Time

Coverage

Compilation Options

Path Count Profiles

Coverage

Count

Low Iteration Count Profiles

Coverage

Count

Experiment Summaries