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-172-31-47-249.ec2.internal/175-802-9624/llama.cpp/run/oneview_runs/multicore/armclang_3/oneview_run_1758031731OMP_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-172-31-47-249.ec2.internal+	1	1	177.21
►Process 8623+		1	177.21
○Thread 8623			177.21	177.76	99.69	0.00	0.00	177.75

▶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-172-31-47-249.ec2.internal/175-802-9624/llama.cpp/run/oneview_runs/multicore/armclang_3/oneview_run_1758031731OMP_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-172-31-47-249.ec2.internal+	1	2	94.73
►Process 8867+		2	94.73
○Thread 8867			94.73	95.42	99.29	0.00	0.00	94.85
○Thread 8966			94.67	94.85	99.81	0.56	0.00	94.82

▶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-172-31-47-249.ec2.internal/175-802-9624/llama.cpp/run/oneview_runs/multicore/armclang_3/oneview_run_1758031731OMP_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-172-31-47-249.ec2.internal+	1	4	49.13
►Process 9048+		4	49.13
○Thread 9048			49.13	49.76	98.73	0.00	0.00	49.20
○Thread 9147			48.99	49.20	99.55	0.55	0.01	49.17
○Thread 9148			49.01	49.20	99.60	0.55	0.01	49.16
○Thread 9149			48.95	49.20	99.48	0.55	0.01	49.17

▶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-172-31-47-249.ec2.internal/175-802-9624/llama.cpp/run/oneview_runs/multicore/armclang_3/oneview_run_1758031731OMP_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-172-31-47-249.ec2.internal+	1	8	27.06
►Process 9236+		8	27.06
○Thread 9236			27.06	27.61	98.00	0.00	0.00	27.05
○Thread 9335			26.87	27.05	99.33	0.55	0.01	27.02
○Thread 9336			26.77	27.05	98.98	0.56	0.01	27.02
○Thread 9337			26.76	27.05	98.92	0.56	0.01	27.02
○Thread 9338			26.83	27.05	99.18	0.56	0.01	27.01
○Thread 9339			26.77	27.05	98.96	0.56	0.01	27.01
○Thread 9340			26.75	27.05	98.90	0.56	0.01	27.02
○Thread 9341			26.76	27.05	98.92	0.56	0.01	27.02

▶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-172-31-47-249.ec2.internal/175-802-9624/llama.cpp/run/oneview_runs/multicore/armclang_3/oneview_run_1758031731OMP_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-172-31-47-249.ec2.internal+	1	16	16.24
►Process 9362+		16	16.24
○Thread 9362			16.24	16.85	96.33	0.00	0.00	16.29
○Thread 9461			16.01	16.29	98.24	0.56	0.01	16.25
○Thread 9462			15.82	16.29	97.07	0.56	0.01	16.26
○Thread 9463			15.91	16.29	97.62	0.56	0.01	16.25
○Thread 9464			16.03	16.29	98.42	0.56	0.01	16.26
○Thread 9465			15.99	16.29	98.15	0.56	0.01	16.26
○Thread 9466			16.04	16.29	98.45	0.56	0.01	16.25
○Thread 9467			16.01	16.29	98.24	0.56	0.01	16.26
○Thread 9468			15.94	16.29	97.81	0.56	0.01	16.26
○Thread 9469			16.01	16.29	98.24	0.56	0.01	16.26
○Thread 9470			15.88	16.29	97.44	0.56	0.01	16.26
○Thread 9471			16.02	16.29	98.30	0.56	0.01	16.26
○Thread 9472			16.05	16.29	98.52	0.56	0.01	16.26
○Thread 9473			15.86	16.29	97.35	0.56	0.01	16.26
○Thread 9474			15.98	16.29	98.09	0.56	0.01	16.26
○Thread 9475			15.97	16.29	98.00	0.56	0.01	16.26

▶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-172-31-47-249.ec2.internal/175-802-9624/llama.cpp/run/oneview_runs/multicore/armclang_3/oneview_run_1758031731OMP_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-172-31-47-249.ec2.internal+	1	24	13.21
►Process 9556+		24	13.21
○Thread 9556			13.21	13.89	95.09	0.00	0.00	13.32
○Thread 9655			13.00	13.33	97.48	0.56	0.01	13.29
○Thread 9656			12.96	13.33	97.22	0.56	0.01	13.29
○Thread 9657			13.14	13.33	98.57	0.56	0.01	13.29
○Thread 9658			13.00	13.33	97.52	0.56	0.01	13.29
○Thread 9659			13.04	13.33	97.78	0.56	0.01	13.29
○Thread 9660			12.98	13.33	97.37	0.56	0.01	13.29
○Thread 9661			13.00	13.33	97.52	0.56	0.01	13.29
○Thread 9662			12.96	13.33	97.19	0.56	0.01	13.29
○Thread 9663			13.06	13.33	97.94	0.56	0.01	13.29
○Thread 9664			13.09	13.33	98.20	0.56	0.01	13.29
○Thread 9665			13.01	13.33	97.60	0.56	0.01	13.29
○Thread 9666			13.08	13.33	98.09	0.56	0.01	13.29
○Thread 9667			13.01	13.33	97.56	0.56	0.01	13.29
○Thread 9668			12.98	13.33	97.34	0.56	0.01	13.29
○Thread 9669			12.96	13.33	97.19	0.56	0.01	13.29
○Thread 9670			13.05	13.33	97.90	0.56	0.01	13.29
○Thread 9671			12.95	13.33	97.15	0.56	0.01	13.29
○Thread 9672			12.91	13.33	96.82	0.56	0.01	13.29
○Thread 9673			12.95	13.33	97.12	0.56	0.01	13.29
○Thread 9674			13.03	13.33	97.76	0.56	0.00	13.29
○Thread 9675			13.02	13.33	97.64	0.56	0.00	13.29
○Thread 9676			13.03	13.33	97.72	0.56	0.00	13.29
○Thread 9677			12.82	13.33	96.18	0.56	0.00	13.29

▶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-172-31-47-249.ec2.internal/175-802-9624/llama.cpp/run/oneview_runs/multicore/armclang_3/oneview_run_1758031731OMP_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-172-31-47-249.ec2.internal+	1	32	12.89
►Process 9698+		32	12.89
○Thread 9698			12.89	13.59	94.85	0.00	0.00	13.00
○Thread 9797			12.69	13.01	97.52	0.58	0.01	12.96
○Thread 9798			12.69	13.01	97.56	0.58	0.01	12.97
○Thread 9799			12.51	13.01	96.13	0.58	0.01	12.97
○Thread 9800			12.62	13.01	97.02	0.58	0.01	12.96
○Thread 9801			12.73	13.01	97.83	0.58	0.01	12.96
○Thread 9802			12.59	13.01	96.79	0.58	0.01	12.97
○Thread 9803			12.65	13.01	97.21	0.58	0.01	12.97
○Thread 9804			12.64	13.01	97.18	0.58	0.01	12.97
○Thread 9805			12.65	13.01	97.21	0.58	0.01	12.97
○Thread 9806			12.68	13.01	97.48	0.58	0.01	12.97
○Thread 9807			12.73	13.01	97.87	0.58	0.01	12.97
○Thread 9808			12.69	13.01	97.60	0.58	0.01	12.95
○Thread 9809			12.77	13.01	98.18	0.58	0.01	12.97
○Thread 9810			12.73	13.01	97.83	0.58	0.01	12.97
○Thread 9811			12.49	13.01	95.99	0.58	0.01	12.97
○Thread 9812			12.71	13.01	97.68	0.58	0.01	12.96
○Thread 9813			12.59	13.01	96.83	0.58	0.01	12.97
○Thread 9814			12.49	13.01	95.99	0.58	0.01	12.97
○Thread 9815			12.62	13.01	97.07	0.58	0.01	12.97
○Thread 9816			12.73	13.01	97.88	0.58	0.01	12.97
○Thread 9817			12.58	13.01	96.72	0.58	0.01	12.97
○Thread 9818			12.67	13.01	97.42	0.58	0.01	12.97
○Thread 9819			12.66	13.01	97.30	0.58	0.01	12.96
○Thread 9820			12.66	13.01	97.34	0.58	0.01	12.97
○Thread 9821			12.75	13.01	97.99	0.58	0.01	12.97
○Thread 9822			12.66	13.01	97.34	0.58	0.01	12.96
○Thread 9823			12.68	13.01	97.46	0.58	0.01	12.97
○Thread 9824			12.66	13.01	97.30	0.58	0.01	12.97
○Thread 9825			12.67	13.01	97.38	0.58	0.01	12.95
○Thread 9826			12.60	13.01	96.84	0.58	0.01	12.97
○Thread 9827			12.65	13.01	97.23	0.58	0.01	12.97

▶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-172-31-47-249.ec2.internal/175-802-9624/llama.cpp/run/oneview_runs/multicore/armclang_3/oneview_run_1758031731OMP_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-172-31-47-249.ec2.internal+	1	40	12.10
►Process 9848+		40	12.10
○Thread 9848			12.10	12.80	94.52	0.00	0.00	12.22
○Thread 9947			11.87	12.23	97.05	0.56	0.01	12.19
○Thread 9948			11.88	12.23	97.13	0.56	0.01	12.17
○Thread 9949			11.91	12.23	97.38	0.56	0.01	12.19
○Thread 9950			11.88	12.23	97.17	0.56	0.01	12.19
○Thread 9951			11.97	12.23	97.87	0.56	0.01	12.19
○Thread 9952			12.01	12.23	98.16	0.56	0.01	12.19
○Thread 9953			11.76	12.23	96.15	0.56	0.01	12.19
○Thread 9954			11.91	12.23	97.38	0.56	0.01	12.19
○Thread 9955			12.01	12.23	98.20	0.56	0.01	12.19
○Thread 9956			11.88	12.23	97.10	0.56	0.01	12.19
○Thread 9957			11.85	12.23	96.85	0.56	0.01	12.19
○Thread 9958			11.93	12.23	97.55	0.56	0.01	12.19
○Thread 9959			11.93	12.23	97.58	0.56	0.01	12.19
○Thread 9960			11.86	12.23	96.93	0.56	0.01	12.19
○Thread 9961			11.81	12.23	96.56	0.56	0.01	12.19
○Thread 9962			11.99	12.23	98.03	0.56	0.01	12.19
○Thread 9963			11.87	12.23	97.05	0.56	0.01	12.18
○Thread 9964			11.83	12.23	96.72	0.56	0.01	12.19
○Thread 9965			11.94	12.23	97.63	0.56	0.01	12.19
○Thread 9966			11.99	12.23	98.03	0.56	0.01	12.19
○Thread 9967			11.85	12.23	96.89	0.56	0.01	12.19
○Thread 9968			11.81	12.23	96.56	0.56	0.01	12.19
○Thread 9969			11.86	12.23	96.93	0.56	0.01	12.19
○Thread 9970			11.92	12.23	97.47	0.56	0.01	12.18
○Thread 9971			11.94	12.23	97.59	0.56	0.01	12.19
○Thread 9972			11.92	12.23	97.47	0.56	0.01	12.19
○Thread 9973			11.97	12.23	97.84	0.56	0.01	12.19
○Thread 9974			11.82	12.23	96.65	0.56	0.01	12.19
○Thread 9975			11.88	12.23	97.14	0.56	0.01	12.18
○Thread 9976			11.97	12.23	97.84	0.56	0.01	12.19
○Thread 9977			11.88	12.23	97.14	0.56	0.01	12.19
○Thread 9978			11.86	12.23	96.94	0.56	0.01	12.19
○Thread 9979			11.91	12.23	97.35	0.56	0.01	12.18
○Thread 9980			11.90	12.23	97.35	0.56	0.01	12.18
○Thread 9981			11.96	12.23	97.76	0.56	0.01	12.19
○Thread 9982			11.90	12.23	97.31	0.56	0.01	12.19
○Thread 9983			11.88	12.23	97.11	0.56	0.01	12.19
○Thread 9984			11.84	12.23	96.78	0.56	0.01	12.19
○Thread 9985			11.78	12.23	96.29	0.56	0.01	12.19

▶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-172-31-47-249.ec2.internal/175-802-9624/llama.cpp/run/oneview_runs/multicore/armclang_3/oneview_run_1758031731OMP_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-172-31-47-249.ec2.internal+	1	48	11.64
►Process 10066+		48	11.64
○Thread 10066			11.64	12.33	94.40	0.00	0.00	11.75
○Thread 10165			11.44	11.77	97.19	0.56	0.01	11.72
○Thread 10166			11.46	11.77	97.40	0.56	0.01	11.72
○Thread 10167			11.41	11.77	96.98	0.56	0.01	11.72
○Thread 10168			11.53	11.77	98.00	0.56	0.01	11.72
○Thread 10169			11.56	11.77	98.21	0.56	0.01	11.71
○Thread 10170			11.30	11.77	96.05	0.56	0.01	11.72
○Thread 10171			11.45	11.77	97.32	0.56	0.01	11.71
○Thread 10172			11.44	11.77	97.19	0.56	0.01	11.72
○Thread 10173			11.52	11.77	97.87	0.56	0.01	11.72
○Thread 10174			11.51	11.76	97.83	0.56	0.01	11.72
○Thread 10175			11.52	11.76	97.88	0.56	0.01	11.71
○Thread 10176			11.51	11.76	97.83	0.56	0.01	11.72
○Thread 10177			11.40	11.76	96.90	0.56	0.01	11.72
○Thread 10178			11.47	11.76	97.45	0.56	0.01	11.72
○Thread 10179			11.52	11.76	97.88	0.56	0.01	11.72
○Thread 10180			11.52	11.76	97.92	0.56	0.01	11.72
○Thread 10181			11.40	11.76	96.86	0.56	0.01	11.72
○Thread 10182			11.40	11.76	96.86	0.56	0.01	11.72
○Thread 10183			11.45	11.76	97.33	0.56	0.01	11.72
○Thread 10184			11.45	11.76	97.29	0.56	0.01	11.71
○Thread 10185			11.40	11.76	96.86	0.56	0.01	11.72
○Thread 10186			11.40	11.76	96.86	0.56	0.01	11.72
○Thread 10187			11.49	11.76	97.63	0.56	0.01	11.72
○Thread 10188			11.45	11.76	97.33	0.56	0.01	11.72
○Thread 10189			11.39	11.76	96.78	0.56	0.01	11.72
○Thread 10190			11.47	11.76	97.50	0.56	0.01	11.71
○Thread 10191			11.41	11.76	96.95	0.56	0.01	11.71
○Thread 10192			11.50	11.76	97.72	0.56	0.01	11.72
○Thread 10193			11.49	11.76	97.68	0.56	0.01	11.72
○Thread 10194			11.50	11.76	97.76	0.56	0.01	11.71
○Thread 10195			11.44	11.76	97.25	0.56	0.01	11.72
○Thread 10196			11.51	11.76	97.81	0.56	0.01	11.72
○Thread 10197			11.32	11.76	96.19	0.56	0.01	11.72
○Thread 10198			11.37	11.76	96.66	0.56	0.01	11.72
○Thread 10199			11.41	11.76	97.00	0.56	0.01	11.71
○Thread 10200			11.47	11.76	97.47	0.56	0.01	11.72
○Thread 10201			11.40	11.76	96.92	0.56	0.01	11.72
○Thread 10202			11.33	11.76	96.28	0.56	0.01	11.72
○Thread 10203			11.44	11.76	97.26	0.56	0.01	11.71
○Thread 10204			11.46	11.76	97.43	0.56	0.01	11.71
○Thread 10205			11.31	11.76	96.16	0.56	0.01	11.72
○Thread 10206			11.39	11.76	96.84	0.56	0.01	11.72
○Thread 10207			11.33	11.76	96.28	0.56	0.01	11.72
○Thread 10208			11.37	11.76	96.63	0.56	0.01	11.71
○Thread 10209			11.45	11.76	97.31	0.56	0.01	11.72
○Thread 10210			11.41	11.76	96.97	0.56	0.01	11.71
○Thread 10211			11.13	11.76	94.59	0.56	0.01	11.72

▶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-172-31-47-249.ec2.internal/175-802-9624/llama.cpp/run/oneview_runs/multicore/armclang_3/oneview_run_1758031731OMP_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-172-31-47-249.ec2.internal+	1	56	11.19
►Process 10232+		56	11.19
○Thread 10232			11.19	11.81	94.73	0.00	0.00	11.23
○Thread 10331			10.96	11.24	97.51	0.56	0.01	11.19
○Thread 10332			11.01	11.24	97.91	0.56	0.01	11.19
○Thread 10333			10.89	11.24	96.84	0.56	0.01	11.19
○Thread 10334			10.99	11.24	97.78	0.56	0.01	11.19
○Thread 10335			11.06	11.24	98.40	0.56	0.01	11.19
○Thread 10336			10.86	11.24	96.58	0.56	0.01	11.19
○Thread 10337			10.89	11.24	96.89	0.56	0.01	11.19
○Thread 10338			11.05	11.24	98.27	0.56	0.01	11.18
○Thread 10339			10.94	11.24	97.34	0.56	0.01	11.19
○Thread 10340			10.95	11.24	97.38	0.56	0.01	11.19
○Thread 10341			10.98	11.24	97.69	0.56	0.01	11.19
○Thread 10342			11.07	11.24	98.49	0.56	0.01	11.19
○Thread 10343			10.91	11.24	97.07	0.56	0.01	11.19
○Thread 10344			10.95	11.24	97.38	0.56	0.01	11.19
○Thread 10345			10.88	11.24	96.81	0.56	0.01	11.18
○Thread 10346			11.04	11.24	98.23	0.56	0.01	11.19
○Thread 10347			10.70	11.24	95.21	0.56	0.01	11.18
○Thread 10348			10.87	11.24	96.68	0.56	0.01	11.19
○Thread 10349			10.84	11.24	96.41	0.56	0.01	11.19
○Thread 10350			10.96	11.24	97.48	0.56	0.01	11.19
○Thread 10351			10.89	11.24	96.86	0.56	0.01	11.19
○Thread 10352			10.85	11.24	96.53	0.56	0.01	11.19
○Thread 10353			10.91	11.24	97.02	0.56	0.01	11.19
○Thread 10354			10.93	11.24	97.25	0.56	0.01	11.19
○Thread 10355			10.93	11.24	97.20	0.56	0.01	11.19
○Thread 10356			10.80	11.24	96.05	0.56	0.01	11.19
○Thread 10357			10.92	11.24	97.16	0.56	0.01	11.19
○Thread 10358			11.05	11.24	98.27	0.56	0.01	11.19
○Thread 10359			10.89	11.24	96.94	0.56	0.01	11.19
○Thread 10360			11.00	11.24	97.83	0.56	0.01	11.19
○Thread 10361			10.94	11.24	97.30	0.56	0.01	11.19
○Thread 10362			10.99	11.24	97.79	0.56	0.01	11.19
○Thread 10363			10.93	11.24	97.21	0.56	0.01	11.19
○Thread 10364			10.95	11.24	97.43	0.56	0.01	11.19
○Thread 10365			10.82	11.24	96.28	0.56	0.01	11.19
○Thread 10366			10.92	11.24	97.17	0.56	0.01	11.19
○Thread 10367			10.88	11.24	96.81	0.56	0.01	11.19
○Thread 10368			10.90	11.24	96.99	0.56	0.01	11.19
○Thread 10369			10.93	11.24	97.26	0.56	0.01	11.19
○Thread 10370			10.91	11.24	97.08	0.56	0.01	11.19
○Thread 10371			10.97	11.24	97.62	0.56	0.01	11.19
○Thread 10372			10.98	11.24	97.71	0.56	0.01	11.19
○Thread 10373			10.84	11.24	96.42	0.56	0.01	11.19
○Thread 10374			10.95	11.24	97.44	0.56	0.01	11.19
○Thread 10375			10.92	11.24	97.17	0.56	0.01	11.19
○Thread 10376			10.97	11.24	97.58	0.56	0.01	11.19
○Thread 10377			10.95	11.24	97.40	0.56	0.01	11.19
○Thread 10378			10.96	11.24	97.53	0.56	0.01	11.19
○Thread 10379			10.98	11.24	97.71	0.56	0.01	11.19
○Thread 10380			10.91	11.24	97.04	0.56	0.01	11.19
○Thread 10381			10.95	11.24	97.45	0.56	0.01	11.19
○Thread 10382			11.01	11.24	97.98	0.56	0.01	11.19
○Thread 10383			10.99	11.24	97.85	0.56	0.01	11.19
○Thread 10384			10.91	11.24	97.09	0.56	0.01	11.19
○Thread 10385			10.65	11.24	94.78	0.56	0.01	11.19

▶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-172-31-47-249.ec2.internal/175-802-9624/llama.cpp/run/oneview_runs/multicore/armclang_3/oneview_run_1758031731OMP_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-172-31-47-249.ec2.internal+	1	64	10.61
►Process 10406+		64	10.61
○Thread 10406			10.61	11.22	94.54	0.00	0.00	10.64
○Thread 10505			10.43	10.65	97.93	0.56	0.01	10.60
○Thread 10506			10.26	10.65	96.29	0.56	0.01	10.60
○Thread 10507			10.37	10.65	97.37	0.56	0.01	10.60
○Thread 10508			10.30	10.65	96.67	0.56	0.01	10.60
○Thread 10509			10.37	10.65	97.42	0.56	0.01	10.60
○Thread 10510			10.36	10.65	97.24	0.56	0.01	10.60
○Thread 10511			10.24	10.65	96.11	0.56	0.01	10.60
○Thread 10512			10.39	10.65	97.55	0.56	0.01	10.60
○Thread 10513			10.34	10.65	97.04	0.56	0.01	10.60
○Thread 10514			10.36	10.65	97.23	0.56	0.01	10.60
○Thread 10515			10.43	10.65	97.93	0.56	0.01	10.60
○Thread 10516			10.39	10.65	97.51	0.56	0.01	10.60
○Thread 10517			10.32	10.65	96.85	0.56	0.01	10.59
○Thread 10518			10.27	10.65	96.48	0.56	0.01	10.60
○Thread 10519			10.37	10.65	97.37	0.56	0.01	10.60
○Thread 10520			10.42	10.65	97.84	0.56	0.01	10.60
○Thread 10521			10.34	10.65	97.05	0.56	0.01	10.60
○Thread 10522			10.38	10.65	97.47	0.56	0.01	10.60
○Thread 10523			10.38	10.65	97.47	0.56	0.01	10.60
○Thread 10524			10.45	10.65	98.08	0.56	0.01	10.60
○Thread 10525			10.39	10.65	97.52	0.56	0.01	10.59
○Thread 10526			10.38	10.65	97.43	0.56	0.01	10.60
○Thread 10527			10.42	10.65	97.85	0.56	0.01	10.60
○Thread 10528			10.37	10.65	97.33	0.56	0.01	10.60
○Thread 10529			10.36	10.65	97.24	0.56	0.01	10.60
○Thread 10530			10.37	10.65	97.38	0.56	0.01	10.60
○Thread 10531			10.36	10.65	97.29	0.56	0.01	10.60
○Thread 10532			10.47	10.65	98.28	0.56	0.01	10.60
○Thread 10533			10.44	10.65	97.99	0.56	0.01	10.60
○Thread 10534			10.32	10.65	96.91	0.56	0.01	10.60
○Thread 10535			10.34	10.65	97.06	0.56	0.01	10.60
○Thread 10536			10.47	10.65	98.28	0.56	0.01	10.60
○Thread 10537			10.35	10.65	97.15	0.56	0.01	10.60
○Thread 10538			10.29	10.65	96.64	0.56	0.01	10.60
○Thread 10539			10.17	10.65	95.56	0.56	0.01	10.60
○Thread 10540			10.40	10.65	97.67	0.56	0.01	10.60
○Thread 10541			10.30	10.65	96.69	0.56	0.01	10.60
○Thread 10542			10.37	10.65	97.34	0.56	0.01	10.60
○Thread 10543			10.35	10.65	97.20	0.56	0.01	10.60
○Thread 10544			10.43	10.65	97.91	0.56	0.01	10.60
○Thread 10545			10.36	10.65	97.30	0.56	0.01	10.60
○Thread 10546			10.40	10.65	97.68	0.56	0.01	10.59
○Thread 10547			10.32	10.65	96.88	0.56	0.01	10.59
○Thread 10548			10.32	10.65	96.88	0.56	0.01	10.60
○Thread 10549			10.26	10.65	96.36	0.56	0.01	10.60
○Thread 10550			10.34	10.65	97.12	0.56	0.01	10.60
○Thread 10551			10.36	10.65	97.31	0.56	0.01	10.60
○Thread 10552			10.48	10.65	98.43	0.56	0.01	10.60
○Thread 10553			10.31	10.65	96.84	0.56	0.01	10.60
○Thread 10554			10.36	10.65	97.26	0.56	0.01	10.60
○Thread 10555			10.40	10.65	97.68	0.56	0.01	10.60
○Thread 10556			10.41	10.65	97.78	0.57	0.01	10.60
○Thread 10557			10.37	10.65	97.40	0.57	0.01	10.60
○Thread 10558			10.37	10.65	97.41	0.57	0.01	10.60
○Thread 10559			10.36	10.65	97.26	0.57	0.01	10.60
○Thread 10560			10.42	10.65	97.83	0.57	0.01	10.60
○Thread 10561			10.24	10.65	96.14	0.57	0.01	10.60
○Thread 10562			10.33	10.65	97.03	0.57	0.01	10.60
○Thread 10563			10.32	10.65	96.89	0.57	0.01	10.60
○Thread 10564			10.40	10.65	97.69	0.57	0.01	10.60
○Thread 10565			10.28	10.65	96.57	0.57	0.01	10.60
○Thread 10566			10.45	10.65	98.12	0.57	0.01	10.60
○Thread 10567			10.44	10.65	98.02	0.57	0.01	10.60

▶Scalability - 1x72

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-172-31-47-249.ec2.internal/175-802-9624/llama.cpp/run/oneview_runs/multicore/armclang_3/oneview_run_1758031731OMP_NUM_THREADS: 72OMP_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-172-31-47-249.ec2.internal+	1	72	10.42
►Process 10647+		72	10.42
○Thread 10647			10.42	11.02	94.50	0.00	0.00	10.44
○Thread 10746			10.14	10.45	96.99	0.56	0.01	10.40
○Thread 10747			10.23	10.45	97.85	0.56	0.01	10.40
○Thread 10748			10.13	10.45	96.90	0.56	0.01	10.40
○Thread 10749			10.17	10.45	97.28	0.56	0.01	10.40
○Thread 10750			10.24	10.45	97.95	0.56	0.01	10.40
○Thread 10751			10.23	10.45	97.81	0.56	0.01	10.40
○Thread 10752			10.14	10.45	96.99	0.56	0.01	10.40
○Thread 10753			10.22	10.45	97.71	0.56	0.01	10.40
○Thread 10754			10.23	10.45	97.86	0.56	0.01	10.40
○Thread 10755			10.15	10.45	97.05	0.56	0.01	10.40
○Thread 10756			10.18	10.45	97.33	0.56	0.01	10.40
○Thread 10757			10.31	10.45	98.58	0.56	0.01	10.40
○Thread 10758			10.18	10.45	97.38	0.56	0.01	10.40
○Thread 10759			10.03	10.45	95.90	0.56	0.01	10.40
○Thread 10760			10.08	10.45	96.43	0.56	0.01	10.40
○Thread 10761			10.26	10.45	98.15	0.56	0.01	10.40
○Thread 10762			10.19	10.45	97.48	0.56	0.01	10.40
○Thread 10763			10.19	10.45	97.48	0.56	0.01	10.40
○Thread 10764			10.20	10.45	97.58	0.56	0.01	10.40
○Thread 10765			10.24	10.45	97.96	0.56	0.01	10.40
○Thread 10766			10.21	10.45	97.68	0.56	0.01	10.40
○Thread 10767			10.11	10.45	96.67	0.56	0.01	10.40
○Thread 10768			10.18	10.45	97.34	0.56	0.01	10.40
○Thread 10769			10.25	10.45	98.01	0.56	0.01	10.40
○Thread 10770			10.19	10.45	97.44	0.56	0.01	10.40
○Thread 10771			10.28	10.45	98.35	0.56	0.01	10.40
○Thread 10772			10.26	10.45	98.11	0.56	0.01	10.40
○Thread 10773			10.28	10.45	98.35	0.56	0.01	10.40
○Thread 10774			10.20	10.45	97.59	0.56	0.01	10.40
○Thread 10775			10.20	10.45	97.54	0.56	0.01	10.40
○Thread 10776			10.10	10.45	96.58	0.56	0.01	10.40
○Thread 10777			10.25	10.45	98.07	0.56	0.01	10.40
○Thread 10778			10.19	10.45	97.45	0.56	0.01	10.40
○Thread 10779			10.18	10.45	97.35	0.56	0.01	10.40
○Thread 10780			10.18	10.45	97.35	0.56	0.01	10.40
○Thread 10781			10.27	10.45	98.22	0.56	0.01	10.40
○Thread 10782			10.19	10.45	97.50	0.56	0.01	10.40
○Thread 10783			10.17	10.45	97.31	0.56	0.01	10.40
○Thread 10784			10.13	10.45	96.93	0.56	0.01	10.40
○Thread 10785			10.25	10.45	98.12	0.56	0.01	10.39
○Thread 10786			10.21	10.45	97.69	0.56	0.01	10.40
○Thread 10787			10.24	10.45	97.93	0.56	0.01	10.40
○Thread 10788			10.09	10.45	96.55	0.56	0.01	10.40
○Thread 10789			10.20	10.45	97.65	0.56	0.01	10.40
○Thread 10790			10.12	10.45	96.79	0.56	0.01	10.40
○Thread 10791			10.19	10.45	97.46	0.56	0.01	10.40
○Thread 10792			10.08	10.45	96.45	0.56	0.01	10.40
○Thread 10793			10.24	10.45	97.99	0.56	0.01	10.40
○Thread 10794			10.07	10.45	96.31	0.56	0.01	10.40
○Thread 10795			10.00	10.45	95.64	0.56	0.01	10.40
○Thread 10796			10.12	10.45	96.84	0.56	0.01	10.40
○Thread 10797			10.19	10.45	97.46	0.57	0.01	10.40
○Thread 10798			10.20	10.45	97.56	0.57	0.01	10.40
○Thread 10799			10.24	10.45	97.94	0.57	0.01	10.40
○Thread 10800			10.24	10.45	98.04	0.57	0.01	10.40
○Thread 10801			10.18	10.45	97.42	0.57	0.01	10.40
○Thread 10802			10.14	10.45	96.99	0.57	0.01	10.40
○Thread 10803			10.20	10.45	97.61	0.57	0.01	10.40
○Thread 10804			10.16	10.45	97.18	0.57	0.01	10.40
○Thread 10805			10.27	10.45	98.28	0.57	0.01	10.40
○Thread 10806			10.22	10.45	97.85	0.57	0.01	10.40
○Thread 10807			10.17	10.45	97.28	0.57	0.01	10.40
○Thread 10808			10.20	10.45	97.62	0.57	0.01	10.40
○Thread 10809			10.21	10.45	97.71	0.57	0.01	10.40
○Thread 10810			10.05	10.45	96.13	0.57	0.01	10.40
○Thread 10811			10.19	10.45	97.52	0.57	0.01	10.40
○Thread 10812			10.19	10.45	97.52	0.57	0.01	10.40
○Thread 10813			10.26	10.45	98.15	0.57	0.01	10.40
○Thread 10814			10.14	10.45	97.00	0.57	0.01	10.40
○Thread 10815			10.14	10.45	97.00	0.57	0.01	10.40
○Thread 10816			9.84	10.45	94.13	0.57	0.01	10.40

▶Scalability - 1x80

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-172-31-47-249.ec2.internal/175-802-9624/llama.cpp/run/oneview_runs/multicore/armclang_3/oneview_run_1758031731OMP_NUM_THREADS: 80OMP_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-172-31-47-249.ec2.internal+	1	80	10.48
►Process 10838+		80	10.48
○Thread 10838			10.48	11.09	94.52	0.00	0.00	10.51
○Thread 10937			10.12	10.52	96.20	0.56	0.01	10.47
○Thread 10938			10.16	10.52	96.53	0.56	0.01	10.47
○Thread 10939			10.16	10.52	96.58	0.56	0.01	10.47
○Thread 10940			10.23	10.52	97.20	0.56	0.01	10.47
○Thread 10941			10.09	10.52	95.91	0.56	0.01	10.46
○Thread 10942			10.18	10.52	96.77	0.56	0.01	10.47
○Thread 10943			10.28	10.52	97.68	0.56	0.01	10.47
○Thread 10944			10.29	10.52	97.82	0.56	0.01	10.47
○Thread 10945			10.27	10.52	97.58	0.56	0.01	10.47
○Thread 10946			10.19	10.52	96.87	0.56	0.01	10.47
○Thread 10947			10.25	10.52	97.39	0.56	0.01	10.47
○Thread 10948			10.24	10.52	97.30	0.56	0.01	10.45
○Thread 10949			10.18	10.52	96.82	0.56	0.01	10.47
○Thread 10950			10.12	10.52	96.16	0.56	0.01	10.47
○Thread 10951			10.14	10.52	96.40	0.56	0.01	10.47
○Thread 10952			10.31	10.52	97.97	0.56	0.01	10.47
○Thread 10953			10.14	10.52	96.35	0.56	0.01	10.46
○Thread 10954			10.16	10.52	96.54	0.56	0.01	10.47
○Thread 10955			10.25	10.52	97.40	0.56	0.01	10.47
○Thread 10956			10.18	10.52	96.74	0.56	0.01	10.47
○Thread 10957			10.23	10.52	97.26	0.56	0.01	10.47
○Thread 10958			10.22	10.52	97.12	0.56	0.01	10.47
○Thread 10959			10.15	10.52	96.50	0.56	0.01	10.47
○Thread 10960			10.18	10.52	96.79	0.56	0.01	10.47
○Thread 10961			10.23	10.52	97.21	0.56	0.01	10.47
○Thread 10962			10.25	10.52	97.40	0.56	0.01	10.47
○Thread 10963			10.24	10.52	97.36	0.56	0.01	10.46
○Thread 10964			10.23	10.52	97.26	0.56	0.01	10.47
○Thread 10965			10.18	10.52	96.79	0.56	0.01	10.46
○Thread 10966			10.22	10.52	97.17	0.56	0.01	10.47
○Thread 10967			10.24	10.52	97.31	0.56	0.01	10.47
○Thread 10968			10.35	10.52	98.35	0.56	0.01	10.47
○Thread 10969			10.13	10.52	96.30	0.56	0.01	10.47
○Thread 10970			10.15	10.52	96.45	0.56	0.01	10.47
○Thread 10971			10.20	10.52	96.92	0.56	0.01	10.47
○Thread 10972			10.11	10.52	96.07	0.56	0.01	10.47
○Thread 10973			10.25	10.52	97.40	0.56	0.01	10.47
○Thread 10974			10.15	10.52	96.55	0.56	0.01	10.47
○Thread 10975			10.22	10.52	97.17	0.56	0.01	10.47
○Thread 10976			10.33	10.52	98.21	0.56	0.01	10.47
○Thread 10977			10.22	10.52	97.12	0.56	0.01	10.47
○Thread 10978			10.28	10.52	97.69	0.56	0.01	10.46
○Thread 10979			10.20	10.52	96.98	0.56	0.01	10.47
○Thread 10980			10.26	10.52	97.55	0.56	0.01	10.47
○Thread 10981			10.25	10.52	97.46	0.56	0.01	10.46
○Thread 10982			10.16	10.52	96.55	0.56	0.01	10.47
○Thread 10983			10.14	10.52	96.41	0.56	0.01	10.47
○Thread 10984			10.31	10.52	98.03	0.56	0.01	10.46
○Thread 10985			10.14	10.52	96.41	0.56	0.01	10.47
○Thread 10986			10.16	10.52	96.65	0.56	0.01	10.47
○Thread 10987			10.16	10.52	96.60	0.56	0.01	10.47
○Thread 10988			10.23	10.52	97.22	0.56	0.01	10.47
○Thread 10989			10.16	10.52	96.61	0.56	0.01	10.46
○Thread 10990			10.21	10.52	97.08	0.56	0.01	10.47
○Thread 10991			10.19	10.52	96.85	0.56	0.01	10.47
○Thread 10992			10.23	10.52	97.23	0.56	0.01	10.47
○Thread 10993			10.28	10.52	97.80	0.56	0.01	10.47
○Thread 10994			10.19	10.52	96.85	0.56	0.01	10.47
○Thread 10995			10.17	10.52	96.71	0.56	0.01	10.47
○Thread 10996			10.28	10.52	97.71	0.56	0.01	10.47
○Thread 10997			10.23	10.52	97.33	0.56	0.01	10.47
○Thread 10998			10.15	10.52	96.52	0.57	0.01	10.47
○Thread 10999			10.29	10.52	97.85	0.57	0.01	10.47
○Thread 11000			10.32	10.52	98.14	0.57	0.01	10.47
○Thread 11001			10.22	10.52	97.23	0.57	0.01	10.46
○Thread 11002			10.17	10.52	96.66	0.57	0.01	10.47
○Thread 11003			10.22	10.52	97.24	0.57	0.01	10.47
○Thread 11004			10.26	10.52	97.52	0.57	0.01	10.47
○Thread 11005			10.19	10.52	96.90	0.57	0.01	10.47
○Thread 11006			10.23	10.52	97.29	0.57	0.01	10.47
○Thread 11007			10.26	10.52	97.52	0.57	0.01	10.47
○Thread 11008			10.33	10.52	98.24	0.57	0.01	10.47
○Thread 11009			10.25	10.52	97.53	0.57	0.01	10.47
○Thread 11010			10.23	10.52	97.29	0.57	0.01	10.47
○Thread 11011			10.21	10.52	97.05	0.57	0.01	10.46
○Thread 11012			10.27	10.51	97.67	0.57	0.01	10.47
○Thread 11013			10.13	10.51	96.34	0.57	0.01	10.46
○Thread 11014			10.17	10.51	96.72	0.57	0.01	10.47
○Thread 11015			9.54	10.51	90.68	0.57	0.01	10.46

▶Scalability - 1x88

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-172-31-47-249.ec2.internal/175-802-9624/llama.cpp/run/oneview_runs/multicore/armclang_3/oneview_run_1758031731OMP_NUM_THREADS: 88OMP_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-172-31-47-249.ec2.internal+	1	88	10.45
►Process 11040+		88	10.45
○Thread 11040			10.45	11.13	93.81	0.00	0.00	10.55
○Thread 11139			10.26	10.57	97.07	0.56	0.01	10.51
○Thread 11140			10.28	10.57	97.22	0.56	0.01	10.52
○Thread 11141			10.16	10.57	96.13	0.56	0.01	10.51
○Thread 11142			10.33	10.57	97.69	0.56	0.01	10.51
○Thread 11143			10.31	10.57	97.50	0.56	0.01	10.51
○Thread 11144			10.31	10.57	97.50	0.56	0.01	10.51
○Thread 11145			10.25	10.57	96.98	0.56	0.01	10.51
○Thread 11146			10.32	10.57	97.60	0.56	0.01	10.51
○Thread 11147			10.22	10.57	96.70	0.56	0.01	10.51
○Thread 11148			10.18	10.57	96.28	0.56	0.01	10.51
○Thread 11149			10.23	10.57	96.75	0.56	0.01	10.51
○Thread 11150			10.32	10.57	97.60	0.56	0.01	10.51
○Thread 11151			10.22	10.57	96.71	0.56	0.01	10.51
○Thread 11152			10.14	10.57	95.95	0.56	0.01	10.51
○Thread 11153			10.27	10.57	97.18	0.56	0.01	10.52
○Thread 11154			10.38	10.57	98.22	0.56	0.01	10.51
○Thread 11155			10.20	10.57	96.52	0.56	0.01	10.52
○Thread 11156			10.25	10.57	96.99	0.56	0.01	10.52
○Thread 11157			10.20	10.57	96.47	0.56	0.01	10.51
○Thread 11158			10.30	10.57	97.47	0.56	0.01	10.51
○Thread 11159			10.23	10.57	96.81	0.56	0.01	10.51
○Thread 11160			10.31	10.57	97.52	0.56	0.01	10.52
○Thread 11161			10.29	10.57	97.38	0.56	0.01	10.52
○Thread 11162			10.31	10.57	97.52	0.56	0.01	10.52
○Thread 11163			10.28	10.57	97.24	0.56	0.01	10.51
○Thread 11164			10.27	10.57	97.19	0.56	0.01	10.52
○Thread 11165			10.22	10.57	96.67	0.56	0.01	10.51
○Thread 11166			10.37	10.57	98.14	0.56	0.01	10.51
○Thread 11167			10.28	10.57	97.24	0.56	0.01	10.52
○Thread 11168			10.31	10.57	97.62	0.56	0.01	10.52
○Thread 11169			10.33	10.57	97.71	0.56	0.01	10.51
○Thread 11170			10.36	10.57	98.05	0.56	0.01	10.52
○Thread 11171			10.18	10.57	96.34	0.56	0.01	10.51
○Thread 11172			10.19	10.57	96.39	0.56	0.01	10.51
○Thread 11173			10.16	10.57	96.11	0.56	0.01	10.52
○Thread 11174			10.27	10.57	97.15	0.56	0.01	10.51
○Thread 11175			10.20	10.57	96.49	0.56	0.01	10.52
○Thread 11176			10.20	10.57	96.49	0.56	0.01	10.51
○Thread 11177			10.27	10.57	97.20	0.56	0.01	10.51
○Thread 11178			10.28	10.57	97.30	0.56	0.01	10.51
○Thread 11179			10.22	10.57	96.68	0.56	0.01	10.51
○Thread 11180			10.25	10.57	97.01	0.56	0.01	10.51
○Thread 11181			10.20	10.57	96.49	0.56	0.01	10.51
○Thread 11182			10.32	10.57	97.72	0.56	0.01	10.51
○Thread 11183			10.23	10.57	96.87	0.56	0.01	10.51
○Thread 11184			10.27	10.57	97.21	0.56	0.01	10.51
○Thread 11185			10.29	10.57	97.44	0.56	0.01	10.52
○Thread 11186			10.35	10.57	97.92	0.56	0.01	10.51
○Thread 11187			10.22	10.56	96.69	0.56	0.01	10.52
○Thread 11188			10.22	10.56	96.74	0.56	0.01	10.52
○Thread 11189			10.13	10.56	95.84	0.56	0.01	10.51
○Thread 11190			10.33	10.56	97.78	0.56	0.01	10.51
○Thread 11191			10.20	10.56	96.55	0.56	0.01	10.52
○Thread 11192			10.28	10.56	97.26	0.56	0.01	10.52
○Thread 11193			10.27	10.56	97.17	0.56	0.01	10.51
○Thread 11194			10.27	10.56	97.21	0.56	0.01	10.50
○Thread 11195			10.22	10.56	96.74	0.56	0.01	10.51
○Thread 11196			10.25	10.56	97.03	0.56	0.01	10.52
○Thread 11197			10.27	10.56	97.22	0.56	0.01	10.52
○Thread 11198			10.31	10.56	97.60	0.56	0.01	10.51
○Thread 11199			10.28	10.56	97.26	0.56	0.01	10.52
○Thread 11200			10.29	10.56	97.36	0.56	0.01	10.52
○Thread 11201			10.27	10.56	97.17	0.56	0.01	10.51
○Thread 11202			10.30	10.56	97.46	0.56	0.01	10.51
○Thread 11203			10.21	10.56	96.65	0.56	0.01	10.51
○Thread 11204			10.13	10.56	95.90	0.56	0.01	10.52
○Thread 11205			10.20	10.56	96.56	0.57	0.01	10.52
○Thread 11206			10.24	10.56	96.89	0.57	0.01	10.51
○Thread 11207			10.17	10.56	96.28	0.57	0.01	10.51
○Thread 11208			10.33	10.56	97.75	0.57	0.01	10.52
○Thread 11209			10.22	10.56	96.70	0.57	0.01	10.50
○Thread 11210			10.35	10.56	98.03	0.57	0.01	10.51
○Thread 11211			10.28	10.56	97.27	0.57	0.01	10.51
○Thread 11212			10.25	10.56	97.04	0.57	0.01	10.52
○Thread 11213			10.26	10.56	97.18	0.57	0.01	10.51
○Thread 11214			10.33	10.56	97.80	0.57	0.01	10.52
○Thread 11215			10.35	10.56	97.94	0.57	0.01	10.50
○Thread 11216			10.31	10.56	97.56	0.57	0.01	10.56
○Thread 11217			10.29	10.56	97.37	0.57	0.01	10.51
○Thread 11218			10.29	10.56	97.42	0.57	0.01	10.52
○Thread 11219			10.25	10.56	97.04	0.57	0.01	10.51
○Thread 11220			10.29	10.56	97.42	0.57	0.01	10.51
○Thread 11221			10.28	10.56	97.28	0.57	0.01	10.52
○Thread 11222			10.27	10.56	97.19	0.57	0.01	10.51
○Thread 11223			10.30	10.56	97.52	0.57	0.01	10.51
○Thread 11224			10.18	10.56	96.43	0.57	0.01	10.51
○Thread 11225			9.84	10.56	93.17	0.57	0.01	10.51

▶Scalability - 1x96

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-172-31-47-249.ec2.internal/175-802-9624/llama.cpp/run/oneview_runs/multicore/armclang_3/oneview_run_1758031731OMP_NUM_THREADS: 96OMP_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-172-31-47-249.ec2.internal+	1	96	10.63
►Process 11307+		96	10.63
○Thread 11307			10.63	11.33	93.75	0.00	0.00	10.74
○Thread 11407			10.37	10.77	96.30	0.56	0.01	10.71
○Thread 11408			10.48	10.77	97.32	0.56	0.01	10.71
○Thread 11409			10.45	10.77	97.00	0.56	0.01	10.71
○Thread 11410			10.47	10.77	97.19	0.56	0.01	10.71
○Thread 11411			10.38	10.77	96.40	0.56	0.01	10.71
○Thread 11412			10.38	10.77	96.40	0.56	0.01	10.71
○Thread 11413			10.47	10.77	97.19	0.56	0.01	10.71
○Thread 11414			10.51	10.77	97.65	0.56	0.01	10.71
○Thread 11415			10.44	10.77	96.91	0.56	0.01	10.71
○Thread 11416			10.42	10.77	96.77	0.56	0.01	10.71
○Thread 11417			10.38	10.77	96.40	0.56	0.01	10.70
○Thread 11418			10.48	10.77	97.33	0.56	0.01	10.71
○Thread 11419			10.39	10.77	96.45	0.56	0.01	10.71
○Thread 11420			10.45	10.77	97.01	0.56	0.01	10.71
○Thread 11421			10.32	10.77	95.80	0.56	0.01	10.71
○Thread 11422			10.53	10.77	97.75	0.56	0.01	10.71
○Thread 11423			10.38	10.77	96.41	0.56	0.01	10.71
○Thread 11424			10.37	10.77	96.32	0.56	0.01	10.71
○Thread 11425			10.44	10.77	96.92	0.56	0.01	10.71
○Thread 11426			10.51	10.77	97.62	0.56	0.01	10.71
○Thread 11427			10.41	10.77	96.69	0.56	0.01	10.71
○Thread 11428			10.46	10.77	97.15	0.56	0.01	10.71
○Thread 11429			10.48	10.77	97.29	0.56	0.01	10.71
○Thread 11430			10.46	10.77	97.16	0.56	0.01	10.71
○Thread 11431			10.43	10.77	96.83	0.56	0.01	10.71
○Thread 11432			10.43	10.77	96.83	0.56	0.01	10.71
○Thread 11433			10.46	10.77	97.20	0.56	0.01	10.71
○Thread 11434			10.50	10.77	97.53	0.56	0.01	10.71
○Thread 11435			10.50	10.77	97.53	0.56	0.01	10.71
○Thread 11436			10.36	10.77	96.23	0.56	0.01	10.70
○Thread 11437			10.43	10.77	96.84	0.56	0.01	10.71
○Thread 11438			10.51	10.77	97.58	0.56	0.01	10.71
○Thread 11439			10.31	10.77	95.82	0.56	0.01	10.71
○Thread 11440			10.40	10.77	96.61	0.56	0.01	10.71
○Thread 11441			10.44	10.77	97.03	0.56	0.01	10.71
○Thread 11442			10.43	10.77	96.89	0.56	0.01	10.70
○Thread 11443			10.41	10.77	96.70	0.56	0.01	10.71
○Thread 11444			10.38	10.77	96.42	0.56	0.01	10.71
○Thread 11445			10.43	10.76	96.84	0.56	0.01	10.71
○Thread 11446			10.41	10.76	96.66	0.56	0.01	10.71
○Thread 11447			10.39	10.76	96.52	0.56	0.01	10.71
○Thread 11448			10.38	10.76	96.43	0.56	0.01	10.71
○Thread 11449			10.37	10.76	96.33	0.56	0.01	10.71
○Thread 11450			10.42	10.76	96.80	0.56	0.01	10.71
○Thread 11451			10.27	10.76	95.45	0.56	0.01	10.71
○Thread 11452			10.32	10.76	95.83	0.56	0.01	10.71
○Thread 11453			10.37	10.76	96.34	0.56	0.01	10.71
○Thread 11454			10.57	10.76	98.20	0.56	0.01	10.71
○Thread 11455			10.43	10.76	96.85	0.56	0.01	10.71
○Thread 11456			10.34	10.76	96.06	0.56	0.01	10.71
○Thread 11457			10.32	10.76	95.83	0.56	0.01	10.71
○Thread 11458			10.41	10.76	96.71	0.56	0.01	10.71
○Thread 11459			10.49	10.76	97.41	0.56	0.01	10.69
○Thread 11460			10.34	10.76	96.02	0.56	0.01	10.71
○Thread 11461			10.44	10.76	96.95	0.56	0.01	10.71
○Thread 11462			10.44	10.76	97.04	0.56	0.01	10.71
○Thread 11463			10.38	10.76	96.44	0.56	0.01	10.71
○Thread 11464			10.34	10.76	96.07	0.56	0.01	10.71
○Thread 11465			10.39	10.76	96.53	0.56	0.01	10.71
○Thread 11466			10.43	10.76	96.91	0.56	0.01	10.70
○Thread 11467			10.48	10.76	97.37	0.56	0.01	10.71
○Thread 11468			10.43	10.76	96.86	0.56	0.01	10.71
○Thread 11469			10.47	10.76	97.28	0.56	0.01	10.71
○Thread 11470			10.59	10.76	98.35	0.56	0.01	10.71
○Thread 11471			10.37	10.76	96.35	0.56	0.01	10.71
○Thread 11472			10.38	10.76	96.44	0.56	0.01	10.71
○Thread 11473			10.36	10.76	96.26	0.56	0.01	10.71
○Thread 11474			10.36	10.76	96.31	0.56	0.01	10.71
○Thread 11475			10.38	10.76	96.45	0.56	0.01	10.71
○Thread 11476			10.36	10.76	96.22	0.56	0.01	10.71
○Thread 11477			10.47	10.76	97.33	0.56	0.01	10.71
○Thread 11478			10.49	10.76	97.43	0.56	0.01	10.71
○Thread 11479			10.41	10.76	96.73	0.56	0.01	10.71
○Thread 11480			10.40	10.76	96.64	0.56	0.01	10.71
○Thread 11481			10.46	10.76	97.19	0.56	0.01	10.71
○Thread 11482			10.55	10.76	97.99	0.57	0.01	10.71
○Thread 11483			10.48	10.76	97.34	0.57	0.01	10.71
○Thread 11484			10.46	10.76	97.15	0.57	0.01	10.71
○Thread 11485			10.44	10.76	96.97	0.57	0.01	10.71
○Thread 11486			10.50	10.76	97.52	0.57	0.01	10.71
○Thread 11487			10.48	10.76	97.39	0.57	0.01	10.71
○Thread 11488			10.36	10.76	96.32	0.57	0.01	10.71
○Thread 11489			10.39	10.76	96.55	0.57	0.01	10.71
○Thread 11490			10.43	10.76	96.92	0.57	0.01	10.71
○Thread 11491			10.39	10.76	96.51	0.57	0.01	10.71
○Thread 11492			10.47	10.76	97.30	0.57	0.01	10.71
○Thread 11493			10.45	10.76	97.16	0.57	0.01	10.71
○Thread 11494			10.46	10.76	97.21	0.57	0.01	10.71
○Thread 11495			10.41	10.76	96.74	0.57	0.01	10.70
○Thread 11496			10.49	10.76	97.44	0.57	0.01	10.71
○Thread 11497			10.40	10.76	96.65	0.57	0.01	10.71
○Thread 11498			10.53	10.76	97.91	0.57	0.01	10.71
○Thread 11499			10.37	10.76	96.33	0.57	0.01	10.71
○Thread 11500			10.31	10.76	95.86	0.57	0.01	10.71
○Thread 11501			9.91	10.76	92.05	0.57	0.01	10.71

Report Configuration