options

Software Topology

Scalability - 1x6

Number processes: 1 Number nodes: 1 Run Command: <executable> -m meta-llama-3.1-8b-instruct-Q8_0.gguf -t <OMP_NUM_THREADS> -b 2048 -ub 512 -npp 128 -ntg 0 -npl 16 -c 16384 --seed 0 --output-format jsonlMPI Command: mpirun -n <number_processes> /usr/bin/numactl -C 0-95Run Directory: .OMP_NUM_THREADS: 6I_MPI_PIN_ORDER: bunchOMP_DISPLAY_AFFINITY: TRUEOMP_PROC_BIND: spreadOMP_AFFINITY_FORMAT: 'OMP: pid %P tid %i thread %n bound to OS proc set {%A}'OMP_DISPLAY_ENV: TRUEI_MPI_PIN_DOMAIN: autoI_MPI_DEBUG: 4OMP_PLACES: threads
IDObserved ProcessesObserved ThreadsTime(s)Elapsed Time (s)Active Time (%)Start (after process) (s)End (before process) (s)Maximum Time on the Same CPU (s)
Node isix06.benchmarkcenter.megware.com+1617.27
Process 7790+617.27
Thread 779017.2734.5250.010.000.0017.51
Thread 789017.1917.5298.0916.990.0217.50
Thread 789117.0217.5297.1516.990.0217.46
Thread 789217.1317.5297.7516.990.0217.49
Thread 789317.1317.5297.7816.990.0217.51
Thread 789417.1517.5297.8716.990.0217.51
×