OV - exec - Topology

Software Topology

▼Scalability - 1x1

Number processes: 1 Number nodes: 1 Run Command: <executable> -m meta-llama-3.1-8b-instruct-Q8_0.gguf -no-cnv -t <OMP_NUM_THREADS> -n 512 -p "what is a LLM?" --seed 0MPI Command: mpirun -n <number_processes> --bind-to none --report-bindings Run Directory: /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-18-66/175-768-6804/llama.cpp/run/oneview_runs/multicore/armclang_4/oneview_run_1757689479OMP_PROC_BIND: spreadOMP_DISPLAY_AFFINITY: TRUEOMP_AFFINITY_FORMAT: 'OMP: pid %P tid %i thread %n bound to OS proc set {%A}'OMP_DISPLAY_ENV: TRUEOMP_PLACES: threadsOMP_NUM_THREADS: 1

ID	Observed Processes	Observed Threads	Time(s)	Elapsed Time (s)	Active Time (%)	Start (after process) (s)	End (before process) (s)	Maximum Time on the Same CPU (s)
►Node ip-172-31-18-66+	1	1	177.65
►Process 2142+		1	177.65
○Thread 2142			177.65	178.45	99.55	0.00	0.00	64.38

▶Scalability - 1x2

Number processes: 1 Number nodes: 1 Run Command: <executable> -m meta-llama-3.1-8b-instruct-Q8_0.gguf -no-cnv -t <OMP_NUM_THREADS> -n 512 -p "what is a LLM?" --seed 0MPI Command: mpirun -n <number_processes> --bind-to none --report-bindings Run Directory: /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-18-66/175-768-6804/llama.cpp/run/oneview_runs/multicore/armclang_4/oneview_run_1757689479OMP_NUM_THREADS: 2OMP_PROC_BIND: spreadOMP_DISPLAY_AFFINITY: TRUEOMP_AFFINITY_FORMAT: 'OMP: pid %P tid %i thread %n bound to OS proc set {%A}'OMP_DISPLAY_ENV: TRUEOMP_PLACES: threads

ID	Observed Processes	Observed Threads	Time(s)	Elapsed Time (s)	Active Time (%)	Start (after process) (s)	End (before process) (s)	Maximum Time on the Same CPU (s)
►Node ip-172-31-18-66+	1	2	96.25
►Process 2243+		2	96.25
○Thread 2243			96.25	97.06	99.17	0.00	0.00	96.29
○Thread 2310			95.87	96.30	99.56	0.76	0.01	96.22

▶Scalability - 1x4

Number processes: 1 Number nodes: 1 Run Command: <executable> -m meta-llama-3.1-8b-instruct-Q8_0.gguf -no-cnv -t <OMP_NUM_THREADS> -n 512 -p "what is a LLM?" --seed 0MPI Command: mpirun -n <number_processes> --bind-to none --report-bindings Run Directory: /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-18-66/175-768-6804/llama.cpp/run/oneview_runs/multicore/armclang_4/oneview_run_1757689479OMP_NUM_THREADS: 4OMP_PROC_BIND: spreadOMP_DISPLAY_AFFINITY: TRUEOMP_AFFINITY_FORMAT: 'OMP: pid %P tid %i thread %n bound to OS proc set {%A}'OMP_DISPLAY_ENV: TRUEOMP_PLACES: threads

ID	Observed Processes	Observed Threads	Time(s)	Elapsed Time (s)	Active Time (%)	Start (after process) (s)	End (before process) (s)	Maximum Time on the Same CPU (s)
►Node ip-172-31-18-66+	1	4	49.78
►Process 2340+		4	49.78
○Thread 2340			49.78	50.63	98.32	0.00	0.00	49.87
○Thread 2407			49.49	49.87	99.22	0.75	0.01	49.81
○Thread 2408			49.53	49.87	99.31	0.75	0.01	49.81
○Thread 2409			49.46	49.87	99.17	0.75	0.01	49.81

▶Scalability - 1x8

Number processes: 1 Number nodes: 1 Run Command: <executable> -m meta-llama-3.1-8b-instruct-Q8_0.gguf -no-cnv -t <OMP_NUM_THREADS> -n 512 -p "what is a LLM?" --seed 0MPI Command: mpirun -n <number_processes> --bind-to none --report-bindings Run Directory: /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-18-66/175-768-6804/llama.cpp/run/oneview_runs/multicore/armclang_4/oneview_run_1757689479OMP_NUM_THREADS: 8OMP_PROC_BIND: spreadOMP_DISPLAY_AFFINITY: TRUEOMP_AFFINITY_FORMAT: 'OMP: pid %P tid %i thread %n bound to OS proc set {%A}'OMP_DISPLAY_ENV: TRUEOMP_PLACES: threads

ID	Observed Processes	Observed Threads	Time(s)	Elapsed Time (s)	Active Time (%)	Start (after process) (s)	End (before process) (s)	Maximum Time on the Same CPU (s)
►Node ip-172-31-18-66+	1	8	27.62
►Process 2437+		8	27.62
○Thread 2437			27.62	28.44	97.10	0.00	0.00	27.67
○Thread 2504			27.20	27.68	98.30	0.76	0.01	27.60
○Thread 2505			27.20	27.68	98.30	0.76	0.01	27.61
○Thread 2506			27.31	27.68	98.70	0.76	0.01	27.59
○Thread 2507			27.15	27.67	98.10	0.76	0.01	27.61
○Thread 2508			27.24	27.67	98.43	0.76	0.01	27.60
○Thread 2509			26.92	27.67	97.29	0.76	0.01	27.61
○Thread 2510			27.28	27.67	98.59	0.76	0.01	27.61

▶Scalability - 1x16

Number processes: 1 Number nodes: 1 Run Command: <executable> -m meta-llama-3.1-8b-instruct-Q8_0.gguf -no-cnv -t <OMP_NUM_THREADS> -n 512 -p "what is a LLM?" --seed 0MPI Command: mpirun -n <number_processes> --bind-to none --report-bindings Run Directory: /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-18-66/175-768-6804/llama.cpp/run/oneview_runs/multicore/armclang_4/oneview_run_1757689479OMP_NUM_THREADS: 16OMP_PROC_BIND: spreadOMP_DISPLAY_AFFINITY: TRUEOMP_AFFINITY_FORMAT: 'OMP: pid %P tid %i thread %n bound to OS proc set {%A}'OMP_DISPLAY_ENV: TRUEOMP_PLACES: threads

ID	Observed Processes	Observed Threads	Time(s)	Elapsed Time (s)	Active Time (%)	Start (after process) (s)	End (before process) (s)	Maximum Time on the Same CPU (s)
►Node ip-172-31-18-66+	1	16	20.47
►Process 2543+		16	20.47
○Thread 2543			20.47	21.42	95.54	0.00	0.00	20.65
○Thread 2610			20.17	20.66	97.64	0.76	0.01	20.57
○Thread 2611			20.18	20.66	97.71	0.76	0.01	20.57
○Thread 2612			20.24	20.66	97.98	0.76	0.01	20.57
○Thread 2613			20.25	20.66	98.03	0.76	0.01	20.57
○Thread 2614			19.81	20.66	95.90	0.76	0.01	20.57
○Thread 2615			20.11	20.66	97.35	0.76	0.01	20.57
○Thread 2616			20.11	20.66	97.38	0.76	0.01	20.56
○Thread 2617			20.21	20.66	97.86	0.76	0.01	20.56
○Thread 2618			20.19	20.66	97.74	0.76	0.01	20.57
○Thread 2619			19.87	20.66	96.22	0.76	0.01	20.57
○Thread 2620			20.11	20.66	97.38	0.76	0.01	20.57
○Thread 2621			20.20	20.66	97.81	0.76	0.01	20.57
○Thread 2622			20.00	20.66	96.85	0.76	0.01	20.57
○Thread 2623			20.20	20.66	97.79	0.76	0.01	20.57
○Thread 2624			20.16	20.66	97.62	0.76	0.01	20.57

▶Scalability - 1x24

Number processes: 1 Number nodes: 1 Run Command: <executable> -m meta-llama-3.1-8b-instruct-Q8_0.gguf -no-cnv -t <OMP_NUM_THREADS> -n 512 -p "what is a LLM?" --seed 0MPI Command: mpirun -n <number_processes> --bind-to none --report-bindings Run Directory: /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-18-66/175-768-6804/llama.cpp/run/oneview_runs/multicore/armclang_4/oneview_run_1757689479OMP_NUM_THREADS: 24OMP_PROC_BIND: spreadOMP_DISPLAY_AFFINITY: TRUEOMP_AFFINITY_FORMAT: 'OMP: pid %P tid %i thread %n bound to OS proc set {%A}'OMP_DISPLAY_ENV: TRUEOMP_PLACES: threads

ID	Observed Processes	Observed Threads	Time(s)	Elapsed Time (s)	Active Time (%)	Start (after process) (s)	End (before process) (s)	Maximum Time on the Same CPU (s)
►Node ip-172-31-18-66+	1	24	18.77
►Process 2651+		24	18.77
○Thread 2651			18.77	19.75	95.08	0.00	0.00	18.97
○Thread 2718			18.42	18.97	97.12	0.77	0.01	18.90
○Thread 2719			18.28	18.97	96.38	0.77	0.01	18.90
○Thread 2720			18.45	18.97	97.25	0.77	0.01	18.90
○Thread 2721			18.46	18.97	97.33	0.77	0.01	18.90
○Thread 2722			18.14	18.97	95.64	0.77	0.01	18.90
○Thread 2723			18.50	18.97	97.51	0.77	0.01	18.90
○Thread 2724			18.32	18.97	96.59	0.77	0.01	18.89
○Thread 2725			18.41	18.97	97.07	0.77	0.01	18.90
○Thread 2726			18.16	18.97	95.75	0.77	0.01	18.90
○Thread 2727			18.19	18.97	95.88	0.77	0.01	18.90
○Thread 2728			18.07	18.97	95.25	0.77	0.01	18.90
○Thread 2729			18.39	18.97	96.94	0.77	0.01	18.90
○Thread 2730			18.48	18.97	97.44	0.77	0.01	18.90
○Thread 2731			18.42	18.97	97.10	0.77	0.01	18.90
○Thread 2732			18.39	18.97	96.94	0.77	0.01	18.90
○Thread 2733			18.44	18.97	97.23	0.77	0.01	18.90
○Thread 2734			18.32	18.97	96.57	0.77	0.01	18.90
○Thread 2735			18.37	18.97	96.83	0.77	0.01	18.90
○Thread 2736			18.23	18.97	96.12	0.77	0.01	18.89
○Thread 2737			18.34	18.97	96.68	0.77	0.01	18.90
○Thread 2738			18.32	18.97	96.60	0.77	0.01	18.90
○Thread 2739			18.35	18.97	96.73	0.77	0.01	18.90
○Thread 2740			18.31	18.97	96.54	0.77	0.01	18.90

▶Scalability - 1x32

Number processes: 1 Number nodes: 1 Run Command: <executable> -m meta-llama-3.1-8b-instruct-Q8_0.gguf -no-cnv -t <OMP_NUM_THREADS> -n 512 -p "what is a LLM?" --seed 0MPI Command: mpirun -n <number_processes> --bind-to none --report-bindings Run Directory: /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-18-66/175-768-6804/llama.cpp/run/oneview_runs/multicore/armclang_4/oneview_run_1757689479OMP_NUM_THREADS: 32OMP_PROC_BIND: spreadOMP_DISPLAY_AFFINITY: TRUEOMP_AFFINITY_FORMAT: 'OMP: pid %P tid %i thread %n bound to OS proc set {%A}'OMP_DISPLAY_ENV: TRUEOMP_PLACES: threads

ID	Observed Processes	Observed Threads	Time(s)	Elapsed Time (s)	Active Time (%)	Start (after process) (s)	End (before process) (s)	Maximum Time on the Same CPU (s)
►Node ip-172-31-18-66+	1	32	17.60
►Process 2767+		32	17.60
○Thread 2767			17.60	18.38	95.78	0.00	0.00	17.60
○Thread 2834			17.18	17.61	97.53	0.76	0.01	17.53
○Thread 2835			17.21	17.61	97.70	0.76	0.01	17.53
○Thread 2836			17.16	17.61	97.45	0.76	0.01	17.54
○Thread 2837			16.99	17.61	96.48	0.76	0.01	17.54
○Thread 2838			16.99	17.61	96.45	0.76	0.01	17.54
○Thread 2839			17.08	17.61	96.97	0.76	0.01	17.54
○Thread 2840			16.99	17.61	96.48	0.76	0.01	17.54
○Thread 2841			17.31	17.61	98.33	0.76	0.01	17.53
○Thread 2842			17.00	17.61	96.54	0.76	0.01	17.54
○Thread 2843			16.88	17.61	95.83	0.76	0.01	17.54
○Thread 2844			17.21	17.61	97.73	0.76	0.01	17.53
○Thread 2845			17.15	17.61	97.39	0.76	0.01	17.54
○Thread 2846			17.21	17.61	97.74	0.76	0.01	17.54
○Thread 2847			17.17	17.61	97.48	0.76	0.01	17.53
○Thread 2848			16.98	17.61	96.40	0.76	0.01	17.54
○Thread 2849			17.05	17.61	96.86	0.76	0.01	17.54
○Thread 2850			17.05	17.61	96.80	0.76	0.01	17.54
○Thread 2851			17.16	17.61	97.43	0.76	0.01	17.52
○Thread 2852			17.04	17.61	96.74	0.76	0.01	17.53
○Thread 2853			17.22	17.61	97.79	0.76	0.01	17.53
○Thread 2854			17.24	17.61	97.91	0.76	0.01	17.54
○Thread 2855			16.96	17.61	96.29	0.76	0.01	17.54
○Thread 2856			17.11	17.61	97.17	0.76	0.01	17.53
○Thread 2857			17.28	17.61	98.14	0.76	0.01	17.54
○Thread 2858			16.88	17.61	95.84	0.76	0.01	17.54
○Thread 2859			16.96	17.61	96.32	0.76	0.01	17.54
○Thread 2860			17.16	17.61	97.43	0.76	0.01	17.54
○Thread 2861			17.06	17.61	96.89	0.76	0.01	17.53
○Thread 2862			17.19	17.61	97.63	0.77	0.01	17.54
○Thread 2863			17.12	17.61	97.23	0.77	0.01	17.54
○Thread 2864			17.14	17.61	97.34	0.77	0.01	17.53

▶Scalability - 1x40

Number processes: 1 Number nodes: 1 Run Command: <executable> -m meta-llama-3.1-8b-instruct-Q8_0.gguf -no-cnv -t <OMP_NUM_THREADS> -n 512 -p "what is a LLM?" --seed 0MPI Command: mpirun -n <number_processes> --bind-to none --report-bindings Run Directory: /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-18-66/175-768-6804/llama.cpp/run/oneview_runs/multicore/armclang_4/oneview_run_1757689479OMP_NUM_THREADS: 40OMP_PROC_BIND: spreadOMP_DISPLAY_AFFINITY: TRUEOMP_AFFINITY_FORMAT: 'OMP: pid %P tid %i thread %n bound to OS proc set {%A}'OMP_DISPLAY_ENV: TRUEOMP_PLACES: threads

ID	Observed Processes	Observed Threads	Time(s)	Elapsed Time (s)	Active Time (%)	Start (after process) (s)	End (before process) (s)	Maximum Time on the Same CPU (s)
►Node ip-172-31-18-66+	1	40	16.58
►Process 2892+		40	16.58
○Thread 2892			16.58	17.41	95.23	0.00	0.00	16.63
○Thread 2959			16.11	16.64	96.83	0.76	0.01	16.56
○Thread 2960			16.18	16.64	97.22	0.76	0.01	16.57
○Thread 2961			16.17	16.64	97.16	0.76	0.01	16.55
○Thread 2962			16.32	16.64	98.06	0.76	0.01	16.56
○Thread 2963			16.22	16.64	97.46	0.76	0.01	16.57
○Thread 2964			16.17	16.64	97.19	0.76	0.01	16.57
○Thread 2965			16.15	16.64	97.07	0.76	0.01	16.56
○Thread 2966			16.23	16.64	97.55	0.76	0.01	16.57
○Thread 2967			16.16	16.64	97.10	0.76	0.01	16.56
○Thread 2968			16.15	16.64	97.07	0.76	0.01	16.56
○Thread 2969			16.14	16.64	97.01	0.76	0.01	16.56
○Thread 2970			16.18	16.64	97.22	0.76	0.01	16.56
○Thread 2971			16.14	16.64	96.98	0.76	0.01	16.56
○Thread 2972			16.10	16.64	96.77	0.76	0.01	16.56
○Thread 2973			16.22	16.64	97.49	0.76	0.01	16.56
○Thread 2974			16.31	16.64	98.01	0.76	0.01	16.56
○Thread 2975			16.15	16.64	97.04	0.76	0.01	16.56
○Thread 2976			16.21	16.64	97.43	0.76	0.01	16.57
○Thread 2977			16.12	16.64	96.86	0.76	0.01	16.56
○Thread 2978			16.19	16.64	97.32	0.76	0.01	16.56
○Thread 2979			16.16	16.64	97.11	0.76	0.01	16.56
○Thread 2980			15.94	16.64	95.81	0.76	0.01	16.56
○Thread 2981			16.24	16.64	97.62	0.76	0.01	16.56
○Thread 2982			16.23	16.64	97.56	0.76	0.01	16.56
○Thread 2983			16.14	16.64	96.99	0.76	0.01	16.57
○Thread 2984			16.14	16.64	97.02	0.76	0.01	16.56
○Thread 2985			16.24	16.64	97.59	0.76	0.01	16.56
○Thread 2986			16.26	16.64	97.71	0.76	0.01	16.55
○Thread 2987			16.12	16.64	96.87	0.76	0.01	16.56
○Thread 2988			16.21	16.64	97.41	0.76	0.01	16.56
○Thread 2989			16.30	16.64	97.95	0.76	0.01	16.56
○Thread 2990			16.17	16.64	97.20	0.76	0.01	16.56
○Thread 2991			16.14	16.64	97.02	0.76	0.01	16.54
○Thread 2992			16.18	16.64	97.23	0.76	0.01	16.56
○Thread 2993			16.13	16.64	96.93	0.76	0.01	16.56
○Thread 2994			16.18	16.64	97.23	0.76	0.01	16.56
○Thread 2995			16.12	16.64	96.87	0.76	0.01	16.56
○Thread 2996			16.09	16.64	96.72	0.76	0.01	16.57
○Thread 2997			15.67	16.64	94.17	0.76	0.01	16.56

▶Scalability - 1x48

Number processes: 1 Number nodes: 1 Run Command: <executable> -m meta-llama-3.1-8b-instruct-Q8_0.gguf -no-cnv -t <OMP_NUM_THREADS> -n 512 -p "what is a LLM?" --seed 0MPI Command: mpirun -n <number_processes> --bind-to none --report-bindings Run Directory: /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-18-66/175-768-6804/llama.cpp/run/oneview_runs/multicore/armclang_4/oneview_run_1757689479OMP_NUM_THREADS: 48OMP_PROC_BIND: spreadOMP_DISPLAY_AFFINITY: TRUEOMP_AFFINITY_FORMAT: 'OMP: pid %P tid %i thread %n bound to OS proc set {%A}'OMP_DISPLAY_ENV: TRUEOMP_PLACES: threads

ID	Observed Processes	Observed Threads	Time(s)	Elapsed Time (s)	Active Time (%)	Start (after process) (s)	End (before process) (s)	Maximum Time on the Same CPU (s)
►Node ip-172-31-18-66+	1	48	15.99
►Process 3024+		48	15.99
○Thread 3024			15.99	16.80	95.17	0.00	0.00	16.00
○Thread 3091			15.50	16.01	96.79	0.78	0.01	15.93
○Thread 3092			15.57	16.01	97.26	0.78	0.01	15.92
○Thread 3093			15.57	16.01	97.23	0.78	0.01	15.93
○Thread 3094			15.52	16.01	96.91	0.78	0.01	15.92
○Thread 3095			15.29	16.01	95.48	0.78	0.01	15.93
○Thread 3096			15.59	16.01	97.38	0.78	0.01	15.93
○Thread 3097			15.37	16.01	96.01	0.78	0.01	15.93
○Thread 3098			15.71	16.01	98.10	0.78	0.01	15.93
○Thread 3099			15.42	16.01	96.29	0.78	0.01	15.93
○Thread 3100			15.44	16.01	96.42	0.78	0.01	15.92
○Thread 3101			15.69	16.01	97.98	0.78	0.01	15.93
○Thread 3102			15.70	16.01	98.04	0.78	0.01	15.93
○Thread 3103			15.53	16.01	97.01	0.78	0.01	15.93
○Thread 3104			15.44	16.01	96.45	0.78	0.01	15.93
○Thread 3105			15.35	16.01	95.86	0.78	0.01	15.93
○Thread 3106			15.64	16.01	97.67	0.78	0.01	15.93
○Thread 3107			15.58	16.01	97.29	0.78	0.01	15.93
○Thread 3108			15.53	16.01	97.01	0.78	0.01	15.93
○Thread 3109			15.61	16.01	97.48	0.78	0.01	15.92
○Thread 3110			15.62	16.01	97.54	0.78	0.01	15.93
○Thread 3111			15.62	16.01	97.54	0.78	0.01	15.93
○Thread 3112			15.59	16.01	97.36	0.78	0.01	15.93
○Thread 3113			15.40	16.01	96.17	0.78	0.01	15.93
○Thread 3114			15.61	16.01	97.48	0.78	0.01	15.93
○Thread 3115			15.52	16.01	96.92	0.78	0.01	15.93
○Thread 3116			15.56	16.01	97.20	0.78	0.01	15.93
○Thread 3117			15.49	16.01	96.74	0.78	0.01	15.93
○Thread 3118			15.52	16.01	96.92	0.78	0.01	15.93
○Thread 3119			15.60	16.01	97.42	0.78	0.01	15.93
○Thread 3120			15.58	16.01	97.30	0.78	0.01	15.93
○Thread 3121			15.48	16.01	96.71	0.78	0.01	15.93
○Thread 3122			15.68	16.01	97.92	0.78	0.01	15.92
○Thread 3123			15.56	16.01	97.17	0.78	0.01	15.93
○Thread 3124			15.32	16.01	95.68	0.78	0.01	15.93
○Thread 3125			15.36	16.01	95.93	0.78	0.01	15.93
○Thread 3126			15.56	16.01	97.21	0.78	0.01	15.93
○Thread 3127			15.38	16.01	96.08	0.78	0.01	15.93
○Thread 3128			15.44	16.01	96.46	0.78	0.01	15.93
○Thread 3129			15.62	16.01	97.55	0.78	0.01	15.93
○Thread 3130			15.58	16.01	97.33	0.78	0.01	15.93
○Thread 3131			15.40	16.01	96.21	0.78	0.01	15.93
○Thread 3132			15.55	16.01	97.12	0.78	0.01	15.93
○Thread 3133			15.32	16.01	95.71	0.78	0.01	15.93
○Thread 3134			15.56	16.01	97.18	0.78	0.01	15.93
○Thread 3135			15.57	16.01	97.27	0.78	0.01	15.93
○Thread 3136			15.58	16.01	97.34	0.78	0.01	15.93
○Thread 3137			14.88	16.01	92.96	0.78	0.01	15.93

▶Scalability - 1x56

Number processes: 1 Number nodes: 1 Run Command: <executable> -m meta-llama-3.1-8b-instruct-Q8_0.gguf -no-cnv -t <OMP_NUM_THREADS> -n 512 -p "what is a LLM?" --seed 0MPI Command: mpirun -n <number_processes> --bind-to none --report-bindings Run Directory: /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-18-66/175-768-6804/llama.cpp/run/oneview_runs/multicore/armclang_4/oneview_run_1757689479OMP_NUM_THREADS: 56OMP_PROC_BIND: spreadOMP_DISPLAY_AFFINITY: TRUEOMP_AFFINITY_FORMAT: 'OMP: pid %P tid %i thread %n bound to OS proc set {%A}'OMP_DISPLAY_ENV: TRUEOMP_PLACES: threads

ID	Observed Processes	Observed Threads	Time(s)	Elapsed Time (s)	Active Time (%)	Start (after process) (s)	End (before process) (s)	Maximum Time on the Same CPU (s)
►Node ip-172-31-18-66+	1	56	16.18
►Process 3164+		56	16.18
○Thread 3164			16.18	17.08	94.71	0.00	0.00	16.29
○Thread 3231			15.92	16.30	97.65	0.76	0.01	16.22
○Thread 3232			15.85	16.30	97.19	0.76	0.01	16.22
○Thread 3233			15.72	16.30	96.42	0.76	0.01	16.22
○Thread 3234			15.98	16.30	97.99	0.76	0.01	16.22
○Thread 3235			15.70	16.30	96.27	0.76	0.01	16.22
○Thread 3236			15.88	16.30	97.40	0.76	0.01	16.22
○Thread 3237			15.72	16.30	96.42	0.76	0.01	16.22
○Thread 3238			15.72	16.30	96.42	0.77	0.01	16.23
○Thread 3239			15.75	16.30	96.61	0.77	0.01	16.23
○Thread 3240			15.80	16.30	96.88	0.77	0.01	16.22
○Thread 3241			15.79	16.30	96.82	0.77	0.01	16.23
○Thread 3242			15.85	16.30	97.19	0.77	0.01	16.23
○Thread 3243			15.74	16.30	96.52	0.77	0.01	16.23
○Thread 3244			15.67	16.30	96.12	0.77	0.01	16.22
○Thread 3245			15.58	16.30	95.57	0.77	0.01	16.22
○Thread 3246			15.80	16.30	96.89	0.77	0.01	16.22
○Thread 3247			15.76	16.30	96.67	0.77	0.01	16.23
○Thread 3248			15.85	16.30	97.22	0.77	0.01	16.23
○Thread 3249			15.67	16.30	96.12	0.77	0.01	16.23
○Thread 3250			15.79	16.30	96.86	0.77	0.01	16.22
○Thread 3251			15.54	16.30	95.32	0.77	0.01	16.22
○Thread 3252			15.78	16.30	96.80	0.77	0.01	16.23
○Thread 3253			15.67	16.30	96.09	0.77	0.01	16.22
○Thread 3254			15.82	16.30	97.04	0.77	0.01	16.22
○Thread 3255			15.89	16.30	97.44	0.77	0.01	16.23
○Thread 3256			15.73	16.30	96.49	0.77	0.01	16.22
○Thread 3257			15.83	16.30	97.11	0.77	0.01	16.23
○Thread 3258			15.76	16.30	96.65	0.77	0.01	16.22
○Thread 3259			15.81	16.30	96.95	0.77	0.01	16.22
○Thread 3260			15.92	16.30	97.66	0.77	0.01	16.22
○Thread 3261			15.80	16.30	96.92	0.77	0.01	16.22
○Thread 3262			15.82	16.30	97.01	0.77	0.01	16.22
○Thread 3263			15.54	16.30	95.33	0.77	0.01	16.22
○Thread 3264			15.63	16.30	95.88	0.77	0.01	16.22
○Thread 3265			15.82	16.30	97.02	0.77	0.01	16.22
○Thread 3266			15.79	16.30	96.83	0.77	0.01	16.22
○Thread 3267			15.57	16.30	95.48	0.77	0.01	16.23
○Thread 3268			15.67	16.30	96.13	0.77	0.01	16.22
○Thread 3269			15.64	16.30	95.94	0.77	0.01	16.23
○Thread 3270			15.80	16.30	96.93	0.77	0.01	16.23
○Thread 3271			15.69	16.30	96.25	0.77	0.01	16.22
○Thread 3272			15.71	16.30	96.34	0.77	0.01	16.22
○Thread 3273			15.74	16.30	96.59	0.77	0.01	16.23
○Thread 3274			15.76	16.30	96.68	0.77	0.01	16.23
○Thread 3275			15.68	16.30	96.16	0.77	0.01	16.22
○Thread 3276			15.75	16.30	96.59	0.77	0.01	16.23
○Thread 3277			15.65	16.30	95.98	0.77	0.01	16.23
○Thread 3278			15.78	16.30	96.77	0.77	0.01	16.22
○Thread 3279			15.74	16.30	96.53	0.77	0.01	16.23
○Thread 3280			15.85	16.30	97.20	0.77	0.01	16.23
○Thread 3281			15.72	16.30	96.41	0.77	0.01	16.22
○Thread 3282			15.73	16.30	96.50	0.77	0.01	16.22
○Thread 3283			15.69	16.30	96.25	0.77	0.01	16.22
○Thread 3284			15.57	16.30	95.52	0.77	0.01	16.23
○Thread 3285			15.12	16.30	92.73	0.77	0.01	16.23

▶Scalability - 1x64

Number processes: 1 Number nodes: 1 Run Command: <executable> -m meta-llama-3.1-8b-instruct-Q8_0.gguf -no-cnv -t <OMP_NUM_THREADS> -n 512 -p "what is a LLM?" --seed 0MPI Command: mpirun -n <number_processes> --bind-to none --report-bindings Run Directory: /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-18-66/175-768-6804/llama.cpp/run/oneview_runs/multicore/armclang_4/oneview_run_1757689479OMP_NUM_THREADS: 64OMP_PROC_BIND: spreadOMP_DISPLAY_AFFINITY: TRUEOMP_AFFINITY_FORMAT: 'OMP: pid %P tid %i thread %n bound to OS proc set {%A}'OMP_DISPLAY_ENV: TRUEOMP_PLACES: threads

ID	Observed Processes	Observed Threads	Time(s)	Elapsed Time (s)	Active Time (%)	Start (after process) (s)	End (before process) (s)	Maximum Time on the Same CPU (s)
►Node ip-172-31-18-66+	1	64	16.11
►Process 3312+		64	16.11
○Thread 3312			16.11	17.04	94.51	0.00	0.00	16.25
○Thread 3379			15.65	16.26	96.24	0.77	0.01	16.16
○Thread 3380			15.64	16.26	96.15	0.77	0.01	16.16
○Thread 3381			15.64	16.26	96.18	0.77	0.01	16.17
○Thread 3382			15.80	16.26	97.13	0.77	0.01	16.16
○Thread 3383			15.72	16.26	96.64	0.77	0.01	16.17
○Thread 3384			15.58	16.26	95.81	0.77	0.01	16.17
○Thread 3385			15.69	16.26	96.46	0.77	0.01	16.16
○Thread 3386			15.69	16.26	96.46	0.77	0.01	16.17
○Thread 3387			15.70	16.26	96.55	0.77	0.01	16.16
○Thread 3388			15.68	16.26	96.40	0.77	0.01	16.17
○Thread 3389			15.62	16.26	96.03	0.77	0.01	16.16
○Thread 3390			15.83	16.26	97.35	0.77	0.01	16.16
○Thread 3391			15.58	16.26	95.78	0.77	0.01	16.17
○Thread 3392			15.77	16.26	96.98	0.77	0.01	16.17
○Thread 3393			15.75	16.26	96.86	0.77	0.01	16.16
○Thread 3394			15.81	16.26	97.23	0.77	0.01	16.16
○Thread 3395			15.57	16.26	95.75	0.77	0.01	16.17
○Thread 3396			15.67	16.26	96.34	0.77	0.01	16.17
○Thread 3397			15.70	16.26	96.55	0.77	0.01	16.17
○Thread 3398			15.66	16.26	96.28	0.77	0.01	16.16
○Thread 3399			15.66	16.26	96.28	0.77	0.01	16.16
○Thread 3400			15.63	16.26	96.12	0.77	0.01	16.16
○Thread 3401			15.73	16.26	96.74	0.77	0.01	16.16
○Thread 3402			15.69	16.26	96.49	0.77	0.01	16.17
○Thread 3403			15.73	16.26	96.74	0.77	0.01	16.17
○Thread 3404			15.73	16.26	96.71	0.77	0.01	16.17
○Thread 3405			15.73	16.26	96.74	0.77	0.01	16.17
○Thread 3406			15.77	16.26	96.99	0.77	0.01	16.17
○Thread 3407			15.66	16.26	96.31	0.77	0.01	16.16
○Thread 3408			15.75	16.26	96.83	0.77	0.01	16.17
○Thread 3409			15.72	16.26	96.68	0.77	0.01	16.17
○Thread 3410			15.83	16.26	97.33	0.77	0.01	16.16
○Thread 3411			15.78	16.26	97.02	0.77	0.01	16.16
○Thread 3412			15.78	16.26	97.05	0.77	0.01	16.16
○Thread 3413			15.68	16.26	96.43	0.77	0.01	16.17
○Thread 3414			15.83	16.26	97.36	0.77	0.01	16.17
○Thread 3415			15.78	16.26	97.02	0.77	0.01	16.17
○Thread 3416			15.66	16.26	96.28	0.77	0.01	16.16
○Thread 3417			15.62	16.26	96.07	0.77	0.01	16.16
○Thread 3418			15.87	16.26	97.58	0.77	0.01	16.16
○Thread 3419			15.70	16.26	96.59	0.77	0.01	16.16
○Thread 3420			15.68	16.26	96.44	0.77	0.01	16.17
○Thread 3421			15.79	16.26	97.12	0.77	0.01	16.17
○Thread 3422			15.82	16.26	97.30	0.77	0.01	16.16
○Thread 3423			15.64	16.26	96.19	0.77	0.01	16.16
○Thread 3424			15.67	16.26	96.35	0.77	0.01	16.17
○Thread 3425			15.73	16.26	96.75	0.77	0.01	16.17
○Thread 3426			15.87	16.26	97.61	0.77	0.01	16.16
○Thread 3427			15.68	16.26	96.41	0.77	0.01	16.17
○Thread 3428			15.58	16.26	95.80	0.77	0.01	16.17
○Thread 3429			15.50	16.26	95.33	0.77	0.01	16.16
○Thread 3430			15.85	16.26	97.46	0.77	0.01	16.16
○Thread 3431			15.67	16.26	96.38	0.77	0.01	16.17
○Thread 3432			15.69	16.26	96.47	0.77	0.01	16.17
○Thread 3433			15.74	16.26	96.78	0.77	0.01	16.16
○Thread 3434			15.83	16.26	97.33	0.77	0.01	16.16
○Thread 3435			15.55	16.26	95.61	0.77	0.01	16.17
○Thread 3436			15.69	16.26	96.51	0.77	0.01	16.17
○Thread 3437			15.68	16.26	96.41	0.77	0.01	16.16
○Thread 3438			15.80	16.26	97.18	0.77	0.01	16.17
○Thread 3439			15.72	16.26	96.66	0.77	0.01	16.17
○Thread 3440			15.73	16.26	96.75	0.77	0.01	16.16
○Thread 3441			15.54	16.26	95.58	0.77	0.01	16.16

Report Configuration