Help is available by moving the cursor above any 
symbol or by checking MAQAO website.
- There is no filter information to display
| Total Time (s) | 4.70 | |
| Max (Thread Active Time) (s) | 3.03 | |
| Average Active Time (s) | 1.91 | |
| Activity Ratio (%) | 75.9 | |
| Average number of active threads | 39.051 | |
| Affinity Stability (%) | 81.5 | |
| Time in analyzed loops (%) | 92.0 | |
| Time in analyzed innermost loops (%) | 89.5 | |
| Time in user code (%) | 92.4 | |
| Compilation Options Score (%) | 100 | |
| Array Access Efficiency (%) | 74.3 | |
|
| Potential Speedups | |
| Perfect Flow Complexity | 1.00 | |
| Perfect OpenMP/MPI/Pthread/TBB | 1.04 | |
| Perfect OpenMP/MPI/Pthread/TBB + Perfect Load Distribution | 1.70 | |
| No Scalar Integer | Potential Speedup | 1.02 | |
| Nb Loops to get 80% | 4 | |
| FP Vectorised | Potential Speedup | 1.01 | |
| Nb Loops to get 80% | 2 | |
| Fully Vectorised | Potential Speedup | 1.69 | |
| Nb Loops to get 80% | 1 | |
| FP Arithmetic Only | Potential Speedup | 3.46 | |
| Nb Loops to get 80% | 1 | |
| Source Object | Issue |
|---|
| ▼libllama.so– | |
| ○hashtable.h | |
| ▼libggml-cpu.so– | |
| ○vec.cpp | |
| ○traits.cpp | |
| ○kai_rhs_pack_nxk_qsi4c32pscalef16_qsu4c32s16s0.c | |
| ○kai_matmul_clamp_f32_qsi8d32p4x8_qsi4c32p4x8_16x4_neon_i8mm.c | |
| ○ops.cpp | |
| ○kai_lhs_quant_pack_qsi8d32p4x8sb_f32_neon.c | |
| ○binary-ops.cpp | |
| ○ggml-cpu.c | |
| ○quants.c | |
| ▼libggml-base.so– | |
| ○ggml.c | |
| Application | /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-35-140.ec2.internal/176-398-1667/llama.cpp/run/binaries/gcc_2/exec |
| Timestamp | 2025-11-24 11:01:41 |
Universal Timestamp | 1763982101 |
| Number of processes observed | 1 |
Number of threads observed | 96 |
| Experiment Type | MPI; OpenMP; |
| Machine | ip-172-31-35-140.ec2.internal |
| Architecture | aarch64 |
Micro Architecture | ARM_NEOVERSE_V2 |
| OS Version | Linux 6.1.158-178.288.amzn2023.aarch64 #1 SMP Mon Nov 3 18:38:05 UTC 2025 |
| Architecture used during static analysis | aarch64 |
Micro Architecture used during static analysis | ARM_NEOVERSE_V2 |
| Frequency Driver | NA |
Frequency Governor | NA |
| Huge Pages | madvise |
Hyperthreading | off |
| Number of sockets | 1 |
Number of cores per socket | 96 |
| Compilation Options | libggml-base.so: GNU C11 11.5.0 20240719 (Red Hat 11.5.0-5) -mcpu=neoverse-v2 -msve-vector-bits=128 -mlittle-endian -mabi=lp64 -g -O3 -O3 -O3 -std=gnu11 -funroll-loops -ffast-math -fno-omit-frame-pointer -fcf-protection=none -fno-finite-math-only -fPIC
libggml-cpu.so: GNU C11 11.5.0 20240719 (Red Hat 11.5.0-5) -mcpu=neoverse-v2 -msve-vector-bits=128 -mlittle-endian -mabi=lp64 -g -O3 -O3 -O3 -std=gnu11 -funroll-loops -ffast-math -fno-omit-frame-pointer -fcf-protection=none -fno-finite-math-only -fPIC -fopenmp
libllama.so: GNU C++17 11.5.0 20240719 (Red Hat 11.5.0-5) -mcpu=neoverse-v2 -msve-vector-bits=128 -mlittle-endian -mabi=lp64 -g -O3 -O3 -O3 -funroll-loops -ffast-math -fno-omit-frame-pointer -fcf-protection=none -fno-finite-math-only -fPIC | | |
| Dataset | |
| Run Command | <executable> -m meta-llama-3.1-8b-instruct-Q4_0.gguf -t 96 -b 2048 -ub 512 -npp 128 -ntg 0 -npl 16 -c 16384 --seed 0 --output-format jsonl |
| MPI Command | mpirun -n <number_processes> --bind-to none --report-bindings /usr/bin/numactl -C 0-95 |
| Number Processes | 1 |
| Number Nodes | 1 |
| Filter | Not Used |
| Profile Start | Not Used |
| Profile Stop | Not Used |
