options

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

Global Metrics

Metricr0r1r2r3r4r5
Total Time (s)50.5162.8568.0768.1068.8569.24
Max (Thread Active Time) (s)42.2554.8060.0059.9560.6961.13
Average Active Time (s)42.2054.3759.5259.4560.1560.57
Activity Ratio (%)97.498.999.099.099.098.9
Average number of active threads6.68455.36083.940111.754139.796167.967
Affinity Stability (%)97.399.599.699.799.799.5
GFLOPS50.92639.58636.18936.24535.83135.604
Time in analyzed loops (%)90.062.162.961.761.561.9
Time in analyzed innermost loops (%)88.962.062.761.661.461.8
Time in user code (%)91.562.463.161.961.862.2
Compilation Options Score (%)99.699.899.899.899.899.8
Array Access Efficiency (%)100100100100100100
Potential Speedups
Perfect Flow Complexity1.001.001.001.001.001.00
Perfect OpenMP/MPI/Pthread/TBB1.051.051.041.031.051.04
Perfect OpenMP/MPI/Pthread/TBB + Perfect Load Distribution1.091.611.591.621.631.62
Scalability - Gap1.009.9516.1721.5727.2632.90
No Scalar IntegerPotential Speedup1.001.001.001.001.001.00
Nb Loops to get 80%111111
FP VectorisedPotential Speedup1.001.001.001.001.001.00
Nb Loops to get 80%111111
Fully VectorisedPotential Speedup1.001.001.001.001.001.00
Nb Loops to get 80%332211
Only FP ArithmeticPotential Speedup1.001.001.001.001.001.00
Nb Loops to get 80%222222
OpenMP perfectly balancedPotential Speedup1.081.611.581.611.611.60
Nb Loops to get 80%111111

FLOPS Breakdown

Scalability Speedup

Cumulated Speedup If No Scalar Integer

Cumulated Speedup If FP Vectorized

Cumulated Speedup If Fully Vectorized

Cumulated Speedup If Only FP Arithmetic

Cumulated Speedup if OpenMP Perfectly Balanced

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

Source ObjectIssue
libllama.so
-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)
-O2, -O3 or -Ofast is missing.
-march=(target) is missing.
exec
-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)
-O2, -O3 or -Ofast is missing.
-march=(target) is missing.
libggml-base.so
-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)
-O2, -O3 or -Ofast is missing.
-march=(target) is missing.
libggml-cpu.so
binary-ops.cpp
ops.cpp
sgemm.cpp
vec.cpp
ggml-cpu.c
quants.c

Path Count Profiles

Coverage

Count

Low Iteration Count Profiles

Coverage

Count

Average Number of Active Threads

Run 1 - 1x8

Run 2 - 1x64

Run 3 - 1x96

Run 4 - 1x128

Run 5 - 1x160

Run 6 - 1x192

Experiment Summaries

r0r1r2r3r4r5
Application/beegfs/hackathon/users/eoseret/qaas_runs_test/175-950-2189/intel/llama.cpp/run/binaries/icx_3/execsame as r0same as r0same as r0same as r0same as r0
Timestamp2025-10-06 13:14:44same as r0same as r0same as r0same as r0same as r0
Experiment TypeMPI; OpenMP; same as r0same as r0same as r0same as r0same as r0
Machinegmz12.benchmarkcenter.megware.comsame as r0same as r0same as r0same as r0same as r0
Architecturex86_64same as r0same as r0same as r0same as r0same as r0
Micro ArchitectureZEN_V5same as r0same as r0same as r0same as r0same as r0
Model NameAMD EPYC 9655 96-Core Processorsame as r0same as r0same as r0same as r0same as r0
Cache Size1024 KBsame as r0same as r0same as r0same as r0same as r0
Number of Cores96same as r0same as r0same as r0same as r0same as r0
Maximal Frequency4.509375 GHzsame as r0same as r0same as r0same as r0same as r0
OS VersionLinux 5.14.0-570.39.1.el9_6.x86_64 #1 SMP PREEMPT_DYNAMIC Thu Sep 4 05:08:52 EDT 2025same as r0same as r0same as r0same as r0same as r0
Architecture used during static analysisx86_64same as r0same as r0same as r0same as r0same as r0
Micro Architecture used during static analysisZEN_V5same as r0same as r0same as r0same as r0same as r0
Compilation Options exec: N/A
libggml-base.so: N/A
libggml-cpu.so: clang based Intel(R) oneAPI DPC++/C++ Compiler 2025.1.0 (2025.1.0.20250317) /cluster/intel/oneapi/2025.1.0/compiler/2025.1/bin/compiler/clang --intel -D GGML_BACKEND_BUILD -D GGML_BACKEND_SHARED -D GGML_SCHED_MAX_COPIES=4 -D GGML_SHARED -D GGML_USE_CPU_REPACK -D GGML_USE_LLAMAFILE -D GGML_USE_OPENMP -D _GNU_SOURCE -D _XOPEN_SOURCE=600 -D ggml_cpu_EXPORTS -I /beegfs/hackathon/users/eoseret/qaas_runs_test/175-950-2189/intel/llama.cpp/build/llama.cpp/ggml/src/.. -I /beegfs/hackathon/users/eoseret/qaas_runs_test/175-950-2189/intel/llama.cpp/build/llama.cpp/ggml/src/. -I /beegfs/hackathon/users/eoseret/qaas_runs_test/175-950-2189/intel/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu -I /beegfs/hackathon/users/eoseret/qaas_runs_test/175-950-2189/intel/llama.cpp/build/llama.cpp/ggml/src/../include -O3 -O3 -march=znver5 -axCORE-AVX512 -mprefer-vector-width=256 -g -fno-omit-frame-pointer -fcf-protection=none -no-pie -grecord-command-line -fno-finite-math-only -O3 -D NDEBUG -std=gnu11 -fPIC -Wshadow -Wstrict-prototypes -Wpointer-arith -Wmissing-prototypes -Werror=implicit-int -Werror=implicit-function-declaration -Wall -Wextra -Wpedantic -Wcast-qual -Wno-unused-function -fno-associative-math -fiopenmp -MD -MT ggml/src/CMakeFiles/ggml-cpu.dir/ggml-cpu/arch/x86/quants.c.o -MF ggml/src/CMakeFiles/ggml-cpu.dir/ggml-cpu/arch/x86/quants.c.o.d -o ggml/src/CMakeFiles/ggml-cpu.dir/ggml-cpu/arch/x86/quants.c.o -c /beegfs/hackathon/users/eoseret/qaas_runs_test/175-950-2189/intel/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/arch/x86/quants.c -fveclib=SVML
libllama.so: N/A
same as r0same as r0same as r0same as r0same as r0
Number of processes observed1same as r0same as r0same as r0same as r0same as r0
Number of threads observed86496128160192
Frequency Driveracpi-cpufreqsame as r0same as r0same as r0same as r0same as r0
Frequency Governorperformancesame as r0same as r0same as r0same as r0same as r0
Huge Pagesalwayssame as r0same as r0same as r0same as r0same as r0
Hyperthreadingonsame as r0same as r0same as r0same as r0same as r0
Number of sockets2same as r0same as r0same as r0same as r0same as r0
Number of cores per socket96same as r0same as r0same as r0same as r0same as r0
MAQAO version2025.1.2same as r0same as r0same as r0same as r0same as r0
MAQAO buildad4b42c12cfbc289a7a711f3ded92abe2eb90c0a::20250917-142411same as r0same as r0same as r0same as r0same as r0
CommentsOV scalability run using icx_3same as r0same as r0same as r0same as r0same as r0
×