OV - exec - Topology

Software Topology

▼Scalability - 1x1

Number processes: 1 Number nodes: 1 Run Command: <executable> -m meta-llama-3.1-8b-instruct-Q4_0.gguf -t <OMP_NUM_THREADS> -b 2048 -ub 512 -npp 0 -ntg 128 -npl 16 -c 16384 --seed 0 --output-format jsonlMPI 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/176-138-2040/llama.cpp/run/oneview_runs/multicore/armclang_3/oneview_run_1761383857OMP_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	172.22
►Process 25174+		1	172.22
○Thread 25174			172.22	173.73	99.13	0.00	0.00	173.72

▶Scalability - 1x2

Number processes: 1 Number nodes: 1 Run Command: <executable> -m meta-llama-3.1-8b-instruct-Q4_0.gguf -t <OMP_NUM_THREADS> -b 2048 -ub 512 -npp 0 -ntg 128 -npl 16 -c 16384 --seed 0 --output-format jsonlMPI 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/176-138-2040/llama.cpp/run/oneview_runs/multicore/armclang_3/oneview_run_1761383857OMP_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	86.71
►Process 25487+		2	86.71
○Thread 25487			86.71	88.29	98.22	0.00	0.00	86.46
○Thread 25586			86.13	86.47	99.61	1.82	0.00	86.25

▶Scalability - 1x4

Number processes: 1 Number nodes: 1 Run Command: <executable> -m meta-llama-3.1-8b-instruct-Q4_0.gguf -t <OMP_NUM_THREADS> -b 2048 -ub 512 -npp 0 -ntg 128 -npl 16 -c 16384 --seed 0 --output-format jsonlMPI 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/176-138-2040/llama.cpp/run/oneview_runs/multicore/armclang_3/oneview_run_1761383857OMP_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	44.00
►Process 25658+		4	44.00
○Thread 25658			44.00	45.51	96.69	0.00	0.00	43.67
○Thread 25757			43.37	43.68	99.29	1.82	0.00	43.46
○Thread 25758			43.37	43.68	99.29	1.82	0.00	43.47
○Thread 25759			43.33	43.68	99.19	1.82	0.00	43.47

▶Scalability - 1x8

Number processes: 1 Number nodes: 1 Run Command: <executable> -m meta-llama-3.1-8b-instruct-Q4_0.gguf -t <OMP_NUM_THREADS> -b 2048 -ub 512 -npp 0 -ntg 128 -npl 16 -c 16384 --seed 0 --output-format jsonlMPI 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/176-138-2040/llama.cpp/run/oneview_runs/multicore/armclang_3/oneview_run_1761383857OMP_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	22.72
►Process 25829+		8	22.72
○Thread 25829			22.72	24.24	93.75	0.00	0.00	22.42
○Thread 25928			22.12	22.43	98.64	1.81	0.00	22.22
○Thread 25929			22.13	22.43	98.66	1.81	0.00	22.22
○Thread 25930			22.15	22.43	98.77	1.81	0.00	22.22
○Thread 25931			22.11	22.43	98.59	1.81	0.00	22.22
○Thread 25932			22.10	22.43	98.55	1.81	0.00	22.22
○Thread 25933			22.10	22.43	98.55	1.81	0.00	22.22
○Thread 25934			22.12	22.43	98.64	1.81	0.00	22.22

▶Scalability - 1x16

Number processes: 1 Number nodes: 1 Run Command: <executable> -m meta-llama-3.1-8b-instruct-Q4_0.gguf -t <OMP_NUM_THREADS> -b 2048 -ub 512 -npp 0 -ntg 128 -npl 16 -c 16384 --seed 0 --output-format jsonlMPI 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/176-138-2040/llama.cpp/run/oneview_runs/multicore/armclang_3/oneview_run_1761383857OMP_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	12.23
►Process 25955+		16	12.23
○Thread 25955			12.23	13.78	88.69	0.00	0.00	11.97
○Thread 26054			11.71	11.97	97.76	1.81	0.01	11.77
○Thread 26055			11.64	11.97	97.21	1.81	0.01	11.76
○Thread 26056			11.64	11.97	97.21	1.81	0.00	11.77
○Thread 26057			11.67	11.97	97.46	1.81	0.00	11.76
○Thread 26058			11.57	11.97	96.58	1.81	0.00	11.76
○Thread 26059			11.54	11.97	96.37	1.81	0.00	11.76
○Thread 26060			11.56	11.97	96.50	1.81	0.00	11.77
○Thread 26061			11.53	11.97	96.29	1.81	0.00	11.76
○Thread 26062			11.60	11.98	96.87	1.81	0.00	11.77
○Thread 26063			11.51	11.98	96.11	1.81	0.00	11.76
○Thread 26064			11.58	11.98	96.70	1.81	0.00	11.76
○Thread 26065			11.55	11.98	96.45	1.81	0.00	11.77
○Thread 26066			11.60	11.98	96.82	1.81	0.00	11.76
○Thread 26067			11.56	11.98	96.48	1.81	0.00	11.76
○Thread 26068			11.52	11.98	96.15	1.81	0.00	11.76

▶Scalability - 1x24

Number processes: 1 Number nodes: 1 Run Command: <executable> -m meta-llama-3.1-8b-instruct-Q4_0.gguf -t <OMP_NUM_THREADS> -b 2048 -ub 512 -npp 0 -ntg 128 -npl 16 -c 16384 --seed 0 --output-format jsonlMPI 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/176-138-2040/llama.cpp/run/oneview_runs/multicore/armclang_3/oneview_run_1761383857OMP_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	9.56
►Process 26089+		24	9.56
○Thread 26089			9.56	11.11	85.99	0.00	0.00	9.28
○Thread 26188			8.95	9.29	96.32	1.82	0.01	9.08
○Thread 26189			8.97	9.29	96.53	1.82	0.01	9.08
○Thread 26190			8.91	9.29	95.88	1.82	0.01	9.08
○Thread 26191			8.88	9.29	95.55	1.82	0.01	9.08
○Thread 26192			8.78	9.29	94.53	1.82	0.01	9.08
○Thread 26193			8.84	9.29	95.12	1.82	0.01	9.08
○Thread 26194			8.82	9.29	94.90	1.82	0.01	9.08
○Thread 26195			8.79	9.29	94.57	1.82	0.01	9.08
○Thread 26196			8.84	9.29	95.11	1.82	0.01	9.08
○Thread 26197			8.79	9.29	94.57	1.82	0.01	9.07
○Thread 26198			8.82	9.29	94.89	1.82	0.00	9.07
○Thread 26199			8.81	9.29	94.78	1.82	0.00	9.08
○Thread 26200			8.75	9.29	94.19	1.82	0.00	9.08
○Thread 26201			8.79	9.29	94.56	1.82	0.00	9.08
○Thread 26202			8.84	9.29	95.15	1.82	0.00	9.07
○Thread 26203			8.78	9.29	94.50	1.82	0.00	9.08
○Thread 26204			8.80	9.29	94.66	1.82	0.00	9.08
○Thread 26205			8.74	9.29	94.01	1.82	0.00	9.08
○Thread 26206			8.80	9.29	94.66	1.82	0.00	9.07
○Thread 26207			8.85	9.29	95.19	1.82	0.00	9.07
○Thread 26208			8.77	9.29	94.33	1.82	0.00	9.07
○Thread 26209			8.77	9.29	94.33	1.82	0.00	9.08
○Thread 26210			8.59	9.29	92.39	1.82	0.00	9.08

▶Scalability - 1x32

Number processes: 1 Number nodes: 1 Run Command: <executable> -m meta-llama-3.1-8b-instruct-Q4_0.gguf -t <OMP_NUM_THREADS> -b 2048 -ub 512 -npp 0 -ntg 128 -npl 16 -c 16384 --seed 0 --output-format jsonlMPI 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/176-138-2040/llama.cpp/run/oneview_runs/multicore/armclang_3/oneview_run_1761383857OMP_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	7.99
►Process 26279+		32	7.99
○Thread 26279			7.99	9.71	82.27	0.00	0.00	7.89
○Thread 26378			7.53	7.89	95.33	1.81	0.01	7.68
○Thread 26379			7.46	7.89	94.51	1.81	0.01	7.68
○Thread 26380			7.46	7.89	94.50	1.81	0.01	7.67
○Thread 26381			7.32	7.89	92.73	1.81	0.01	7.67
○Thread 26382			7.32	7.89	92.66	1.81	0.01	7.68
○Thread 26383			7.30	7.89	92.47	1.81	0.01	7.68
○Thread 26384			7.25	7.89	91.83	1.81	0.01	7.68
○Thread 26385			7.28	7.89	92.15	1.81	0.01	7.68
○Thread 26386			7.26	7.90	91.89	1.81	0.01	7.68
○Thread 26387			7.25	7.90	91.76	1.81	0.01	7.69
○Thread 26388			7.21	7.90	91.32	1.81	0.01	7.68
○Thread 26389			7.22	7.90	91.38	1.81	0.01	7.69
○Thread 26390			7.29	7.90	92.33	1.81	0.01	7.68
○Thread 26391			7.24	7.90	91.63	1.81	0.01	7.68
○Thread 26392			7.22	7.90	91.37	1.81	0.01	7.68
○Thread 26393			7.29	7.90	92.25	1.81	0.01	7.69
○Thread 26394			7.21	7.90	91.30	1.81	0.01	7.68
○Thread 26395			7.23	7.90	91.55	1.81	0.01	7.68
○Thread 26396			7.24	7.90	91.61	1.81	0.00	7.68
○Thread 26397			7.28	7.90	92.12	1.81	0.00	7.67
○Thread 26398			7.19	7.90	91.04	1.81	0.00	7.68
○Thread 26399			7.31	7.90	92.56	1.81	0.00	7.68
○Thread 26400			7.24	7.90	91.67	1.81	0.00	7.68
○Thread 26401			7.24	7.90	91.60	1.81	0.00	7.67
○Thread 26402			7.23	7.90	91.53	1.81	0.00	7.68
○Thread 26403			7.23	7.90	91.53	1.81	0.00	7.67
○Thread 26404			7.26	7.90	91.91	1.81	0.00	7.68
○Thread 26405			7.29	7.90	92.22	1.81	0.00	7.67
○Thread 26406			7.21	7.90	91.27	1.81	0.00	7.68
○Thread 26407			7.33	7.90	92.79	1.81	0.00	7.67
○Thread 26408			7.31	7.90	92.53	1.81	0.00	7.68

▶Scalability - 1x40

Number processes: 1 Number nodes: 1 Run Command: <executable> -m meta-llama-3.1-8b-instruct-Q4_0.gguf -t <OMP_NUM_THREADS> -b 2048 -ub 512 -npp 0 -ntg 128 -npl 16 -c 16384 --seed 0 --output-format jsonlMPI 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/176-138-2040/llama.cpp/run/oneview_runs/multicore/armclang_3/oneview_run_1761383857OMP_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	6.91
►Process 26429+		40	6.91
○Thread 26429			6.91	8.74	78.96	0.00	0.00	6.90
○Thread 26528			6.54	6.92	94.48	1.81	0.01	6.71
○Thread 26529			6.55	6.92	94.69	1.81	0.01	6.71
○Thread 26530			6.54	6.92	94.47	1.81	0.01	6.71
○Thread 26531			6.40	6.92	92.52	1.81	0.01	6.71
○Thread 26532			6.37	6.92	92.08	1.81	0.01	6.70
○Thread 26533			6.33	6.92	91.50	1.81	0.01	6.70
○Thread 26534			6.28	6.92	90.70	1.81	0.01	6.70
○Thread 26535			6.34	6.92	91.64	1.81	0.01	6.71
○Thread 26536			6.23	6.92	89.97	1.81	0.01	6.71
○Thread 26537			6.24	6.92	90.12	1.82	0.01	6.70
○Thread 26538			6.25	6.92	90.33	1.82	0.01	6.70
○Thread 26539			6.32	6.92	91.34	1.82	0.01	6.71
○Thread 26540			6.24	6.92	90.11	1.82	0.01	6.69
○Thread 26541			6.25	6.92	90.32	1.82	0.01	6.70
○Thread 26542			6.23	6.92	90.03	1.82	0.01	6.71
○Thread 26543			6.25	6.92	90.31	1.82	0.01	6.71
○Thread 26544			6.16	6.92	88.94	1.82	0.01	6.70
○Thread 26545			6.20	6.92	89.51	1.82	0.01	6.70
○Thread 26546			6.19	6.92	89.37	1.82	0.01	6.71
○Thread 26547			6.31	6.92	91.10	1.82	0.01	6.70
○Thread 26548			6.28	6.92	90.73	1.82	0.01	6.69
○Thread 26549			6.23	6.92	89.93	1.82	0.01	6.71
○Thread 26550			6.26	6.92	90.36	1.82	0.01	6.70
○Thread 26551			6.26	6.92	90.36	1.82	0.01	6.71
○Thread 26552			6.21	6.92	89.64	1.82	0.01	6.71
○Thread 26553			6.22	6.92	89.78	1.82	0.01	6.71
○Thread 26554			6.21	6.92	89.70	1.82	0.00	6.70
○Thread 26555			6.24	6.92	90.13	1.82	0.00	6.70
○Thread 26556			6.20	6.92	89.48	1.82	0.00	6.71
○Thread 26557			6.24	6.92	90.05	1.82	0.00	6.71
○Thread 26558			6.26	6.92	90.34	1.82	0.00	6.70
○Thread 26559			6.22	6.92	89.83	1.82	0.00	6.70
○Thread 26560			6.24	6.92	90.04	1.82	0.00	6.70
○Thread 26561			6.23	6.92	89.97	1.82	0.00	6.70
○Thread 26562			6.19	6.92	89.32	1.82	0.00	6.70
○Thread 26563			6.27	6.93	90.47	1.82	0.00	6.71
○Thread 26564			6.19	6.93	89.31	1.82	0.00	6.71
○Thread 26565			6.26	6.93	90.32	1.82	0.00	6.69
○Thread 26566			6.01	6.93	86.78	1.82	0.00	6.71

▶Scalability - 1x48

Number processes: 1 Number nodes: 1 Run Command: <executable> -m meta-llama-3.1-8b-instruct-Q4_0.gguf -t <OMP_NUM_THREADS> -b 2048 -ub 512 -npp 0 -ntg 128 -npl 16 -c 16384 --seed 0 --output-format jsonlMPI 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/176-138-2040/llama.cpp/run/oneview_runs/multicore/armclang_3/oneview_run_1761383857OMP_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	6.46
►Process 26589+		48	6.46
○Thread 26589			6.46	8.23	78.52	0.00	0.00	6.38
○Thread 26690			6.03	6.40	94.16	1.81	0.02	6.19
○Thread 26691			5.98	6.40	93.37	1.81	0.02	6.18
○Thread 26692			6.04	6.40	94.31	1.81	0.02	6.19
○Thread 26693			5.93	6.40	92.67	1.81	0.02	6.19
○Thread 26694			5.81	6.40	90.79	1.81	0.02	6.18
○Thread 26695			5.73	6.40	89.53	1.81	0.02	6.18
○Thread 26696			5.73	6.40	89.53	1.81	0.01	6.18
○Thread 26697			5.61	6.40	87.65	1.81	0.01	6.19
○Thread 26698			5.69	6.40	88.82	1.81	0.01	6.19
○Thread 26699			5.64	6.40	88.11	1.81	0.01	6.19
○Thread 26700			5.61	6.40	87.64	1.81	0.01	6.19
○Thread 26701			5.67	6.40	88.58	1.81	0.01	6.19
○Thread 26702			5.58	6.40	87.09	1.81	0.01	6.19
○Thread 26703			5.71	6.40	89.12	1.81	0.01	6.19
○Thread 26704			5.65	6.40	88.18	1.81	0.01	6.18
○Thread 26705			5.60	6.40	87.47	1.81	0.01	6.19
○Thread 26706			5.63	6.40	87.86	1.81	0.01	6.19
○Thread 26707			5.58	6.40	87.15	1.81	0.01	6.19
○Thread 26708			5.61	6.40	87.62	1.81	0.01	6.18
○Thread 26709			5.57	6.40	86.91	1.81	0.01	6.18
○Thread 26710			5.66	6.40	88.32	1.81	0.01	6.19
○Thread 26711			5.58	6.40	87.14	1.81	0.01	6.18
○Thread 26712			5.54	6.40	86.51	1.81	0.01	6.18
○Thread 26713			5.57	6.40	86.98	1.81	0.01	6.19
○Thread 26714			5.64	6.40	87.99	1.81	0.01	6.18
○Thread 26715			5.54	6.40	86.51	1.81	0.01	6.18
○Thread 26716			5.55	6.40	86.58	1.81	0.01	6.19
○Thread 26717			5.66	6.40	88.37	1.82	0.01	6.18
○Thread 26718			5.55	6.40	86.57	1.82	0.01	6.19
○Thread 26719			5.54	6.41	86.49	1.82	0.01	6.18
○Thread 26720			5.71	6.41	89.07	1.82	0.01	6.19
○Thread 26721			5.68	6.41	88.59	1.82	0.01	6.19
○Thread 26722			5.57	6.41	86.95	1.82	0.01	6.18
○Thread 26723			5.60	6.41	87.42	1.82	0.01	6.19
○Thread 26724			5.64	6.41	87.96	1.82	0.00	6.18
○Thread 26725			5.58	6.41	87.10	1.82	0.00	6.19
○Thread 26726			5.65	6.41	88.11	1.82	0.00	6.18
○Thread 26727			5.59	6.41	87.17	1.82	0.00	6.18
○Thread 26728			5.57	6.41	86.93	1.82	0.00	6.18
○Thread 26729			5.63	6.41	87.79	1.82	0.00	6.19
○Thread 26730			5.66	6.41	88.25	1.82	0.00	6.18
○Thread 26731			5.59	6.41	87.16	1.82	0.00	6.19
○Thread 26732			5.65	6.41	88.09	1.82	0.00	6.18
○Thread 26733			5.53	6.41	86.29	1.82	0.00	6.18
○Thread 26734			5.55	6.41	86.60	1.82	0.00	6.19
○Thread 26735			5.48	6.41	85.51	1.82	0.00	6.18
○Thread 26736			5.43	6.41	84.65	1.82	0.00	6.18

▶Scalability - 1x56

Number processes: 1 Number nodes: 1 Run Command: <executable> -m meta-llama-3.1-8b-instruct-Q4_0.gguf -t <OMP_NUM_THREADS> -b 2048 -ub 512 -npp 0 -ntg 128 -npl 16 -c 16384 --seed 0 --output-format jsonlMPI 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/176-138-2040/llama.cpp/run/oneview_runs/multicore/armclang_3/oneview_run_1761383857OMP_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	5.95
►Process 26807+		56	5.95
○Thread 26807			5.95	7.85	75.79	0.00	0.00	5.99
○Thread 26906			5.67	6.01	94.19	1.82	0.02	5.80
○Thread 26907			5.62	6.01	93.35	1.82	0.02	5.80
○Thread 26908			5.62	6.02	93.35	1.82	0.02	5.80
○Thread 26909			5.59	6.02	92.85	1.82	0.02	5.80
○Thread 26910			5.57	6.02	92.59	1.82	0.02	5.80
○Thread 26911			5.53	6.02	91.92	1.82	0.02	5.80
○Thread 26912			5.49	6.02	91.17	1.82	0.02	5.80
○Thread 26913			5.58	6.02	92.67	1.82	0.02	5.80
○Thread 26914			5.33	6.02	88.59	1.82	0.02	5.80
○Thread 26915			5.21	6.02	86.59	1.82	0.02	5.80
○Thread 26916			5.25	6.02	87.25	1.82	0.02	5.80
○Thread 26917			5.19	6.02	86.17	1.82	0.02	5.80
○Thread 26918			5.13	6.02	85.17	1.82	0.02	5.79
○Thread 26919			5.17	6.02	85.91	1.82	0.02	5.80
○Thread 26920			5.18	6.02	85.99	1.82	0.01	5.80
○Thread 26921			5.25	6.02	87.15	1.82	0.01	5.80
○Thread 26922			5.17	6.02	85.82	1.82	0.01	5.80
○Thread 26923			5.18	6.02	86.07	1.82	0.01	5.80
○Thread 26924			5.19	6.02	86.15	1.82	0.01	5.80
○Thread 26925			5.17	6.02	85.81	1.82	0.01	5.79
○Thread 26926			5.12	6.02	84.98	1.82	0.01	5.80
○Thread 26927			5.14	6.02	85.39	1.82	0.01	5.80
○Thread 26928			5.12	6.02	84.97	1.82	0.01	5.80
○Thread 26929			5.14	6.02	85.30	1.82	0.01	5.80
○Thread 26930			5.05	6.02	83.80	1.82	0.01	5.80
○Thread 26931			5.10	6.02	84.71	1.82	0.01	5.80
○Thread 26932			5.15	6.02	85.46	1.82	0.01	5.80
○Thread 26933			5.09	6.02	84.54	1.82	0.01	5.80
○Thread 26934			5.12	6.02	85.04	1.82	0.01	5.80
○Thread 26935			5.17	6.02	85.78	1.82	0.01	5.80
○Thread 26936			5.19	6.02	86.11	1.82	0.01	5.80
○Thread 26937			5.25	6.02	87.10	1.82	0.01	5.80
○Thread 26938			5.16	6.02	85.60	1.82	0.01	5.80
○Thread 26939			5.13	6.02	85.10	1.82	0.01	5.80
○Thread 26940			5.05	6.02	83.85	1.82	0.01	5.80
○Thread 26941			5.13	6.02	85.10	1.82	0.01	5.80
○Thread 26942			5.16	6.02	85.59	1.82	0.01	5.80
○Thread 26943			5.13	6.02	85.09	1.82	0.01	5.80
○Thread 26944			5.08	6.02	84.26	1.82	0.01	5.80
○Thread 26945			5.17	6.02	85.83	1.82	0.01	5.80
○Thread 26946			5.09	6.02	84.50	1.82	0.01	5.80
○Thread 26947			5.24	6.02	86.90	1.82	0.01	5.80
○Thread 26948			5.17	6.02	85.82	1.82	0.00	5.80
○Thread 26949			5.21	6.02	86.48	1.82	0.00	5.80
○Thread 26950			5.03	6.02	83.49	1.82	0.00	5.80
○Thread 26951			5.17	6.02	85.73	1.82	0.00	5.80
○Thread 26952			5.21	6.03	86.39	1.82	0.00	5.80
○Thread 26953			5.15	6.03	85.47	1.82	0.00	5.80
○Thread 26954			5.03	6.03	83.39	1.82	0.00	5.80
○Thread 26955			5.14	6.03	85.22	1.82	0.00	5.80
○Thread 26956			5.15	6.03	85.46	1.82	0.00	5.80
○Thread 26957			5.06	6.03	83.97	1.82	0.00	5.80
○Thread 26958			5.07	6.03	84.13	1.82	0.00	5.80
○Thread 26959			5.06	6.03	83.88	1.82	0.00	5.80
○Thread 26960			4.94	6.03	81.88	1.82	0.00	5.80

▶Scalability - 1x64

Number processes: 1 Number nodes: 1 Run Command: <executable> -m meta-llama-3.1-8b-instruct-Q4_0.gguf -t <OMP_NUM_THREADS> -b 2048 -ub 512 -npp 0 -ntg 128 -npl 16 -c 16384 --seed 0 --output-format jsonlMPI 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/176-138-2040/llama.cpp/run/oneview_runs/multicore/armclang_3/oneview_run_1761383857OMP_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	5.48
►Process 26981+		64	5.48
○Thread 26981			5.48	7.54	72.64	0.00	0.00	5.69
○Thread 27080			5.38	5.71	94.20	1.81	0.02	5.50
○Thread 27081			5.30	5.71	92.88	1.81	0.02	5.49
○Thread 27082			5.29	5.71	92.61	1.81	0.02	5.49
○Thread 27083			5.24	5.71	91.73	1.81	0.02	5.50
○Thread 27084			5.11	5.71	89.54	1.81	0.02	5.49
○Thread 27085			5.01	5.71	87.78	1.81	0.02	5.49
○Thread 27086			4.75	5.71	83.31	1.81	0.02	5.49
○Thread 27087			5.09	5.71	89.09	1.81	0.02	5.49
○Thread 27088			4.88	5.71	85.41	1.81	0.02	5.49
○Thread 27089			4.94	5.71	86.45	1.81	0.02	5.48
○Thread 27090			4.84	5.71	84.70	1.81	0.02	5.49
○Thread 27091			4.88	5.71	85.48	1.81	0.02	5.49
○Thread 27092			4.91	5.71	86.09	1.81	0.02	5.49
○Thread 27093			4.89	5.71	85.56	1.81	0.02	5.48
○Thread 27094			4.79	5.71	83.90	1.81	0.02	5.49
○Thread 27095			4.77	5.71	83.63	1.81	0.02	5.49
○Thread 27096			4.75	5.71	83.19	1.81	0.02	5.49
○Thread 27097			4.77	5.71	83.62	1.81	0.02	5.48
○Thread 27098			4.80	5.71	84.06	1.81	0.02	5.48
○Thread 27099			4.75	5.71	83.27	1.81	0.02	5.49
○Thread 27100			4.74	5.71	82.91	1.81	0.02	5.48
○Thread 27101			4.81	5.71	84.22	1.81	0.02	5.49
○Thread 27102			4.83	5.71	84.48	1.81	0.02	5.48
○Thread 27103			4.67	5.71	81.85	1.81	0.02	5.48
○Thread 27104			4.74	5.71	82.90	1.81	0.02	5.49
○Thread 27105			4.69	5.71	82.11	1.81	0.01	5.49
○Thread 27106			4.60	5.71	80.53	1.81	0.01	5.49
○Thread 27107			4.71	5.71	82.54	1.81	0.01	5.48
○Thread 27108			4.68	5.71	81.92	1.81	0.01	5.49
○Thread 27109			4.77	5.71	83.50	1.81	0.01	5.47
○Thread 27110			4.69	5.71	82.09	1.81	0.01	5.48
○Thread 27111			4.73	5.71	82.88	1.81	0.01	5.49
○Thread 27112			4.76	5.71	83.22	1.81	0.01	5.49
○Thread 27113			4.85	5.71	84.87	1.81	0.01	5.49
○Thread 27114			4.73	5.71	82.77	1.81	0.01	5.48
○Thread 27115			4.66	5.71	81.54	1.81	0.01	5.48
○Thread 27116			4.67	5.72	81.63	1.81	0.01	5.49
○Thread 27117			4.80	5.72	83.90	1.81	0.01	5.48
○Thread 27118			4.81	5.72	84.07	1.81	0.01	5.49
○Thread 27119			4.83	5.72	84.50	1.81	0.01	5.48
○Thread 27120			4.71	5.72	82.40	1.81	0.01	5.49
○Thread 27121			4.71	5.72	82.40	1.81	0.01	5.48
○Thread 27122			4.69	5.72	82.04	1.81	0.01	5.49
○Thread 27123			4.76	5.72	83.35	1.81	0.01	5.49
○Thread 27124			4.75	5.72	83.17	1.81	0.01	5.48
○Thread 27125			4.77	5.72	83.43	1.81	0.01	5.48
○Thread 27126			4.71	5.72	82.38	1.81	0.01	5.48
○Thread 27127			4.78	5.72	83.60	1.81	0.01	5.48
○Thread 27128			4.77	5.72	83.42	1.81	0.01	5.49
○Thread 27129			4.72	5.72	82.54	1.81	0.01	5.49
○Thread 27130			4.79	5.72	83.76	1.81	0.00	5.48
○Thread 27131			4.67	5.72	81.66	1.81	0.00	5.49
○Thread 27132			4.72	5.72	82.53	1.81	0.00	5.49
○Thread 27133			4.67	5.72	81.65	1.81	0.00	5.49
○Thread 27134			4.71	5.72	82.26	1.81	0.00	5.49
○Thread 27135			4.87	5.72	85.14	1.81	0.00	5.48
○Thread 27136			4.80	5.72	83.83	1.81	0.00	5.49
○Thread 27137			4.75	5.72	83.13	1.81	0.00	5.48
○Thread 27138			4.75	5.72	83.12	1.81	0.00	5.49
○Thread 27139			4.78	5.72	83.64	1.81	0.00	5.49
○Thread 27140			4.70	5.72	82.15	1.81	0.00	5.48
○Thread 27141			4.74	5.72	82.94	1.81	0.00	5.49
○Thread 27142			4.79	5.72	83.81	1.81	0.00	5.48

▶Scalability - 1x72

Number processes: 1 Number nodes: 1 Run Command: <executable> -m meta-llama-3.1-8b-instruct-Q4_0.gguf -t <OMP_NUM_THREADS> -b 2048 -ub 512 -npp 0 -ntg 128 -npl 16 -c 16384 --seed 0 --output-format jsonlMPI 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/176-138-2040/llama.cpp/run/oneview_runs/multicore/armclang_3/oneview_run_1761383857OMP_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	5.26
►Process 27163+		72	5.26
○Thread 27163			5.26	7.37	71.31	0.00	0.00	5.52
○Thread 27262			5.09	5.53	92.03	1.81	0.03	5.31
○Thread 27263			5.08	5.53	91.76	1.81	0.02	5.32
○Thread 27264			5.08	5.53	91.84	1.81	0.02	5.31
○Thread 27265			5.13	5.53	92.74	1.81	0.02	5.31
○Thread 27266			4.99	5.53	90.21	1.81	0.02	5.31
○Thread 27267			5.02	5.53	90.66	1.81	0.02	5.31
○Thread 27268			4.69	5.53	84.78	1.81	0.02	5.31
○Thread 27269			5.05	5.53	91.19	1.81	0.02	5.30
○Thread 27270			4.84	5.53	87.39	1.81	0.02	5.31
○Thread 27271			4.83	5.53	87.30	1.81	0.02	5.31
○Thread 27272			4.66	5.53	84.13	1.81	0.02	5.31
○Thread 27273			4.66	5.53	84.22	1.82	0.02	5.31
○Thread 27274			4.69	5.53	84.76	1.82	0.02	5.31
○Thread 27275			4.64	5.53	83.85	1.82	0.02	5.30
○Thread 27276			4.56	5.53	82.40	1.82	0.02	5.30
○Thread 27277			4.68	5.53	84.66	1.82	0.02	5.31
○Thread 27278			4.57	5.53	82.67	1.82	0.02	5.30
○Thread 27279			4.56	5.53	82.48	1.82	0.02	5.31
○Thread 27280			4.48	5.53	81.03	1.82	0.02	5.31
○Thread 27281			4.53	5.53	81.93	1.82	0.02	5.31
○Thread 27282			4.52	5.54	81.75	1.82	0.02	5.31
○Thread 27283			4.49	5.54	81.11	1.82	0.02	5.31
○Thread 27284			4.71	5.54	84.99	1.82	0.02	5.31
○Thread 27285			4.52	5.54	81.74	1.82	0.02	5.31
○Thread 27286			4.51	5.54	81.56	1.82	0.02	5.31
○Thread 27287			4.48	5.54	81.01	1.82	0.02	5.30
○Thread 27288			4.59	5.54	82.99	1.82	0.02	5.31
○Thread 27289			4.60	5.54	83.08	1.82	0.02	5.31
○Thread 27290			4.54	5.54	82.08	1.82	0.02	5.30
○Thread 27291			4.55	5.54	82.26	1.82	0.01	5.30
○Thread 27292			4.51	5.54	81.54	1.82	0.01	5.30
○Thread 27293			4.55	5.54	82.16	1.82	0.01	5.30
○Thread 27294			4.50	5.54	81.26	1.82	0.01	5.30
○Thread 27295			4.48	5.54	80.89	1.82	0.01	5.31
○Thread 27296			4.56	5.54	82.34	1.82	0.01	5.30
○Thread 27297			4.52	5.54	81.70	1.82	0.01	5.31
○Thread 27298			4.41	5.54	79.62	1.82	0.01	5.30
○Thread 27299			4.47	5.54	80.70	1.82	0.01	5.30
○Thread 27300			4.59	5.54	82.96	1.82	0.01	5.30
○Thread 27301			4.55	5.54	82.14	1.82	0.01	5.31
○Thread 27302			4.52	5.54	81.59	1.82	0.01	5.30
○Thread 27303			4.36	5.54	78.70	1.82	0.01	5.30
○Thread 27304			4.53	5.54	81.86	1.82	0.01	5.30
○Thread 27305			4.53	5.54	81.77	1.82	0.01	5.30
○Thread 27306			4.45	5.54	80.41	1.82	0.01	5.30
○Thread 27307			4.52	5.54	81.67	1.82	0.01	5.30
○Thread 27308			4.53	5.54	81.85	1.82	0.01	5.30
○Thread 27309			4.61	5.54	83.29	1.82	0.01	5.31
○Thread 27310			4.32	5.54	77.96	1.82	0.01	5.30
○Thread 27311			4.55	5.54	82.20	1.82	0.01	5.31
○Thread 27312			4.54	5.54	81.92	1.82	0.01	5.30
○Thread 27313			4.49	5.54	81.02	1.82	0.01	5.30
○Thread 27314			4.56	5.54	82.28	1.82	0.01	5.30
○Thread 27315			4.47	5.54	80.65	1.82	0.01	5.30
○Thread 27316			4.54	5.54	82.00	1.82	0.01	5.31
○Thread 27317			4.63	5.54	83.53	1.82	0.01	5.30
○Thread 27318			4.53	5.54	81.72	1.82	0.01	5.30
○Thread 27319			4.56	5.54	82.35	1.82	0.01	5.30
○Thread 27320			4.46	5.54	80.54	1.82	0.00	5.30
○Thread 27321			4.58	5.54	82.61	1.82	0.00	5.30
○Thread 27322			4.51	5.54	81.44	1.82	0.00	5.30
○Thread 27323			4.53	5.54	81.70	1.82	0.00	5.30
○Thread 27324			4.54	5.54	81.88	1.82	0.00	5.30
○Thread 27325			4.52	5.54	81.52	1.82	0.00	5.31
○Thread 27326			4.46	5.55	80.52	1.82	0.00	5.31
○Thread 27327			4.50	5.55	81.15	1.82	0.00	5.30
○Thread 27328			4.37	5.55	78.89	1.82	0.00	5.31
○Thread 27329			4.34	5.55	78.35	1.82	0.00	5.30
○Thread 27330			4.34	5.55	78.26	1.82	0.00	5.30
○Thread 27331			4.24	5.55	76.45	1.82	0.00	5.30
○Thread 27332			4.30	5.55	77.62	1.82	0.00	5.30

▶Scalability - 1x80

Number processes: 1 Number nodes: 1 Run Command: <executable> -m meta-llama-3.1-8b-instruct-Q4_0.gguf -t <OMP_NUM_THREADS> -b 2048 -ub 512 -npp 0 -ntg 128 -npl 16 -c 16384 --seed 0 --output-format jsonlMPI 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/176-138-2040/llama.cpp/run/oneview_runs/multicore/armclang_3/oneview_run_1761383857OMP_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	5.03
►Process 27353+		80	5.03
○Thread 27353			5.03	7.31	68.82	0.00	0.00	5.44
○Thread 27452			4.91	5.47	89.74	1.81	0.03	5.25
○Thread 27453			4.97	5.47	90.83	1.81	0.03	5.25
○Thread 27454			4.98	5.47	91.01	1.81	0.03	5.25
○Thread 27455			4.91	5.47	89.73	1.81	0.03	5.26
○Thread 27456			4.53	5.47	82.78	1.81	0.03	5.24
○Thread 27457			4.57	5.47	83.60	1.81	0.03	5.25
○Thread 27458			4.41	5.47	80.67	1.81	0.03	5.24
○Thread 27459			4.51	5.47	82.40	1.81	0.03	5.25
○Thread 27460			4.45	5.47	81.30	1.81	0.03	5.25
○Thread 27461			4.40	5.47	80.39	1.81	0.02	5.24
○Thread 27462			4.45	5.47	81.30	1.81	0.02	5.25
○Thread 27463			4.42	5.47	80.84	1.81	0.02	5.26
○Thread 27464			4.38	5.47	80.10	1.81	0.02	5.23
○Thread 27465			4.33	5.47	79.10	1.81	0.02	5.26
○Thread 27466			4.40	5.47	80.46	1.81	0.02	5.26
○Thread 27467			4.39	5.47	80.19	1.81	0.02	5.25
○Thread 27468			4.42	5.48	80.73	1.81	0.02	5.25
○Thread 27469			4.26	5.48	77.80	1.81	0.02	5.25
○Thread 27470			4.36	5.48	79.63	1.81	0.02	5.26
○Thread 27471			4.32	5.48	78.99	1.81	0.02	5.25
○Thread 27472			4.40	5.48	80.44	1.81	0.02	5.25
○Thread 27473			4.32	5.48	78.89	1.81	0.02	5.25
○Thread 27474			4.27	5.48	77.97	1.81	0.02	5.25
○Thread 27475			4.36	5.48	79.61	1.81	0.02	5.25
○Thread 27476			4.32	5.48	78.88	1.81	0.02	5.25
○Thread 27477			4.26	5.48	77.87	1.81	0.02	5.24
○Thread 27478			4.27	5.48	78.05	1.81	0.02	5.25
○Thread 27479			4.24	5.48	77.41	1.81	0.02	5.25
○Thread 27480			4.37	5.48	79.78	1.81	0.02	5.25
○Thread 27481			4.31	5.48	78.68	1.81	0.02	5.24
○Thread 27482			4.12	5.48	75.21	1.81	0.02	5.25
○Thread 27483			4.35	5.48	79.49	1.81	0.02	5.24
○Thread 27484			4.24	5.48	77.48	1.81	0.02	5.25
○Thread 27485			4.29	5.48	78.30	1.81	0.02	5.24
○Thread 27486			4.43	5.48	80.85	1.81	0.02	5.25
○Thread 27487			4.35	5.48	79.48	1.81	0.02	5.24
○Thread 27488			4.26	5.48	77.74	1.81	0.02	5.25
○Thread 27489			4.24	5.48	77.47	1.81	0.01	5.25
○Thread 27490			4.31	5.48	78.74	1.81	0.01	5.25
○Thread 27491			4.36	5.48	79.65	1.81	0.01	5.25
○Thread 27492			4.28	5.48	78.10	1.81	0.01	5.25
○Thread 27493			4.36	5.48	79.65	1.82	0.01	5.25
○Thread 27494			4.30	5.48	78.55	1.82	0.01	5.23
○Thread 27495			4.17	5.48	76.08	1.82	0.01	5.25
○Thread 27496			4.39	5.48	80.09	1.82	0.01	5.25
○Thread 27497			4.31	5.48	78.72	1.82	0.01	5.20
○Thread 27498			4.29	5.48	78.26	1.82	0.01	5.23
○Thread 27499			4.29	5.48	78.26	1.82	0.01	5.25
○Thread 27500			4.30	5.48	78.44	1.82	0.01	5.25
○Thread 27501			4.35	5.48	79.35	1.82	0.01	5.25
○Thread 27502			4.32	5.48	78.80	1.82	0.01	5.23
○Thread 27503			4.25	5.48	77.52	1.82	0.01	5.25
○Thread 27504			4.24	5.48	77.33	1.82	0.01	5.25
○Thread 27505			4.25	5.48	77.60	1.82	0.01	5.25
○Thread 27506			4.26	5.48	77.78	1.82	0.01	5.23
○Thread 27507			4.27	5.48	77.96	1.82	0.01	5.25
○Thread 27508			4.31	5.48	78.59	1.82	0.01	5.24
○Thread 27509			4.36	5.48	79.59	1.82	0.01	5.25
○Thread 27510			4.34	5.48	79.23	1.82	0.01	5.25
○Thread 27511			4.25	5.48	77.49	1.82	0.01	5.24
○Thread 27512			4.37	5.48	79.67	1.82	0.01	5.26
○Thread 27513			4.33	5.49	78.94	1.82	0.01	5.26
○Thread 27514			4.34	5.49	79.12	1.82	0.01	5.25
○Thread 27515			4.29	5.49	78.30	1.82	0.01	5.25
○Thread 27516			4.28	5.49	78.02	1.82	0.01	5.23
○Thread 27517			4.27	5.49	77.84	1.82	0.01	5.23
○Thread 27518			4.24	5.49	77.38	1.82	0.00	5.24
○Thread 27519			4.29	5.49	78.19	1.82	0.00	5.25
○Thread 27520			4.22	5.49	76.91	1.82	0.00	5.25
○Thread 27521			4.25	5.49	77.55	1.82	0.00	5.25
○Thread 27522			4.31	5.49	78.55	1.82	0.00	5.25
○Thread 27523			4.22	5.49	76.99	1.82	0.00	5.25
○Thread 27524			4.20	5.49	76.63	1.82	0.00	5.25
○Thread 27525			4.28	5.49	78.08	1.82	0.00	5.26
○Thread 27526			4.28	5.49	77.99	1.82	0.00	5.24
○Thread 27527			4.18	5.49	76.16	1.82	0.00	5.25
○Thread 27528			4.21	5.49	76.70	1.82	0.00	5.25
○Thread 27529			4.28	5.49	77.98	1.82	0.00	5.26
○Thread 27530			3.92	5.49	71.51	1.82	0.00	5.24

▶Scalability - 1x88

Number processes: 1 Number nodes: 1 Run Command: <executable> -m meta-llama-3.1-8b-instruct-Q4_0.gguf -t <OMP_NUM_THREADS> -b 2048 -ub 512 -npp 0 -ntg 128 -npl 16 -c 16384 --seed 0 --output-format jsonlMPI 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/176-138-2040/llama.cpp/run/oneview_runs/multicore/armclang_3/oneview_run_1761383857OMP_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	5.15
►Process 27599+		88	5.15
○Thread 27599			5.15	7.28	70.63	0.00	0.00	5.43
○Thread 27698			5.04	5.46	92.37	1.80	0.03	5.24
○Thread 27699			4.91	5.46	89.99	1.80	0.03	5.24
○Thread 27700			4.95	5.46	90.62	1.80	0.03	5.24
○Thread 27701			4.78	5.46	87.60	1.80	0.03	5.24
○Thread 27702			4.54	5.46	83.29	1.80	0.03	5.24
○Thread 27703			4.46	5.46	81.82	1.80	0.03	5.24
○Thread 27704			4.39	5.46	80.44	1.80	0.03	5.24
○Thread 27705			4.51	5.46	82.63	1.80	0.03	5.23
○Thread 27706			4.42	5.46	81.07	1.80	0.03	5.23
○Thread 27707			4.42	5.46	81.07	1.80	0.03	5.23
○Thread 27708			4.29	5.46	78.68	1.80	0.03	5.24
○Thread 27709			4.38	5.46	80.24	1.80	0.03	5.24
○Thread 27710			4.31	5.46	79.04	1.80	0.03	5.23
○Thread 27711			4.24	5.46	77.67	1.80	0.03	5.23
○Thread 27712			4.37	5.46	80.05	1.80	0.03	5.24
○Thread 27713			4.30	5.46	78.85	1.80	0.03	5.24
○Thread 27714			4.32	5.46	79.21	1.80	0.03	5.20
○Thread 27715			4.23	5.46	77.47	1.80	0.03	5.24
○Thread 27716			4.35	5.46	79.67	1.80	0.03	5.24
○Thread 27717			4.24	5.46	77.74	1.80	0.03	5.23
○Thread 27718			4.14	5.46	75.91	1.80	0.03	5.24
○Thread 27719			4.21	5.46	77.09	1.80	0.02	5.22
○Thread 27720			4.22	5.46	77.36	1.80	0.02	5.24
○Thread 27721			4.39	5.46	80.47	1.80	0.02	5.24
○Thread 27722			4.30	5.46	78.82	1.80	0.02	5.24
○Thread 27723			4.27	5.46	78.18	1.80	0.02	5.24
○Thread 27724			4.20	5.46	76.98	1.80	0.02	5.23
○Thread 27725			4.25	5.46	77.90	1.80	0.02	5.24
○Thread 27726			4.15	5.46	75.97	1.80	0.02	5.24
○Thread 27727			4.20	5.46	76.88	1.80	0.02	5.23
○Thread 27728			4.25	5.46	77.89	1.80	0.02	5.24
○Thread 27729			4.25	5.46	77.79	1.80	0.02	5.23
○Thread 27730			4.26	5.46	78.06	1.80	0.02	5.24
○Thread 27731			4.33	5.46	79.34	1.80	0.02	5.24
○Thread 27732			4.38	5.46	80.16	1.80	0.02	5.23
○Thread 27733			4.27	5.46	78.14	1.80	0.02	5.23
○Thread 27734			4.19	5.46	76.68	1.80	0.02	5.23
○Thread 27735			4.20	5.46	76.86	1.80	0.02	5.23
○Thread 27736			4.15	5.46	75.94	1.80	0.02	5.24
○Thread 27737			4.22	5.47	77.22	1.80	0.02	5.22
○Thread 27738			4.25	5.47	77.76	1.80	0.02	5.24
○Thread 27739			4.16	5.47	76.20	1.80	0.02	5.23
○Thread 27740			4.16	5.47	76.11	1.80	0.02	5.23
○Thread 27741			4.27	5.47	78.12	1.80	0.02	5.23
○Thread 27742			4.15	5.47	75.92	1.80	0.02	5.24
○Thread 27743			4.19	5.47	76.74	1.80	0.02	5.24
○Thread 27744			4.29	5.47	78.56	1.80	0.02	5.24
○Thread 27745			4.31	5.47	78.93	1.80	0.02	5.24
○Thread 27746			4.19	5.47	76.64	1.80	0.02	5.24
○Thread 27747			4.29	5.47	78.55	1.80	0.01	5.23
○Thread 27748			4.23	5.47	77.45	1.80	0.01	5.24
○Thread 27749			4.34	5.47	79.46	1.80	0.01	5.23
○Thread 27750			4.21	5.47	77.08	1.80	0.01	5.24
○Thread 27751			4.15	5.47	75.98	1.80	0.01	5.24
○Thread 27752			4.26	5.47	77.99	1.80	0.01	5.23
○Thread 27753			4.32	5.47	78.99	1.80	0.01	5.22
○Thread 27754			4.21	5.47	77.07	1.80	0.01	5.23
○Thread 27755			4.08	5.47	74.59	1.80	0.01	5.22
○Thread 27756			4.21	5.47	76.97	1.80	0.01	5.23
○Thread 27757			4.14	5.47	75.78	1.80	0.01	5.23
○Thread 27758			4.16	5.47	76.05	1.80	0.01	5.24
○Thread 27759			4.30	5.47	78.60	1.80	0.01	5.24
○Thread 27760			4.27	5.47	78.05	1.80	0.01	5.24
○Thread 27761			4.27	5.47	78.05	1.80	0.01	5.24
○Thread 27762			4.24	5.47	77.49	1.80	0.01	5.24
○Thread 27763			4.26	5.47	77.94	1.80	0.01	5.24
○Thread 27764			4.12	5.47	75.38	1.80	0.01	5.24
○Thread 27765			4.22	5.47	77.20	1.80	0.01	5.24
○Thread 27766			4.32	5.47	78.93	1.80	0.01	5.24
○Thread 27767			4.25	5.47	77.74	1.80	0.01	5.23
○Thread 27768			4.22	5.47	77.19	1.80	0.01	5.22
○Thread 27769			4.21	5.47	76.91	1.80	0.01	5.23
○Thread 27770			4.11	5.47	75.18	1.80	0.01	5.24
○Thread 27771			4.28	5.47	78.19	1.81	0.01	5.23
○Thread 27772			4.19	5.47	76.54	1.81	0.00	5.23
○Thread 27773			4.25	5.47	77.72	1.81	0.00	5.24
○Thread 27774			4.33	5.47	79.18	1.81	0.00	5.23
○Thread 27775			4.20	5.48	76.71	1.81	0.00	5.23
○Thread 27776			4.26	5.48	77.89	1.81	0.00	5.22
○Thread 27777			4.33	5.48	79.17	1.81	0.00	5.23
○Thread 27778			4.26	5.48	77.80	1.81	0.00	5.24
○Thread 27779			4.24	5.48	77.43	1.81	0.00	5.24
○Thread 27780			4.20	5.48	76.78	1.81	0.00	5.23
○Thread 27781			4.30	5.48	78.52	1.81	0.00	5.24
○Thread 27782			4.09	5.48	74.68	1.81	0.00	5.22
○Thread 27783			4.05	5.48	74.04	1.81	0.00	5.23
○Thread 27784			4.05	5.48	73.94	1.81	0.00	5.23

▶Scalability - 1x96

Number processes: 1 Number nodes: 1 Run Command: <executable> -m meta-llama-3.1-8b-instruct-Q4_0.gguf -t <OMP_NUM_THREADS> -b 2048 -ub 512 -npp 0 -ntg 128 -npl 16 -c 16384 --seed 0 --output-format jsonlMPI 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/176-138-2040/llama.cpp/run/oneview_runs/multicore/armclang_3/oneview_run_1761383857OMP_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	5.00
►Process 27806+		96	5.00
○Thread 27806			4.85	7.32	66.31	0.00	0.00	5.43
○Thread 27905			5.00	5.46	91.56	1.83	0.03	5.23
○Thread 27906			4.99	5.46	91.38	1.83	0.03	5.24
○Thread 27907			4.78	5.46	87.53	1.83	0.03	5.23
○Thread 27908			4.75	5.46	86.98	1.83	0.03	5.23
○Thread 27909			4.43	5.46	81.11	1.83	0.03	5.24
○Thread 27910			4.36	5.46	79.92	1.83	0.03	5.22
○Thread 27911			4.39	5.46	80.37	1.83	0.03	5.23
○Thread 27912			4.46	5.46	81.65	1.83	0.03	5.24
○Thread 27913			4.28	5.46	78.45	1.83	0.03	5.23
○Thread 27914			4.24	5.46	77.62	1.83	0.03	5.24
○Thread 27915			4.36	5.46	79.81	1.83	0.03	5.22
○Thread 27916			4.37	5.46	79.99	1.83	0.03	5.20
○Thread 27917			4.14	5.46	75.87	1.83	0.03	5.24
○Thread 27918			4.26	5.46	77.97	1.83	0.03	5.24
○Thread 27919			4.26	5.46	78.06	1.83	0.03	5.23
○Thread 27920			4.46	5.46	81.63	1.83	0.03	5.23
○Thread 27921			4.29	5.46	78.60	1.83	0.03	5.22
○Thread 27922			4.08	5.46	74.76	1.83	0.03	5.23
○Thread 27923			4.20	5.46	76.95	1.83	0.03	5.22
○Thread 27924			4.22	5.46	77.22	1.83	0.03	5.23
○Thread 27925			4.25	5.47	77.86	1.83	0.03	5.23
○Thread 27926			4.16	5.47	76.21	1.83	0.03	5.22
○Thread 27927			4.14	5.47	75.75	1.83	0.03	5.23
○Thread 27928			4.16	5.47	76.11	1.83	0.03	5.23
○Thread 27929			4.16	5.47	76.20	1.83	0.03	5.20
○Thread 27930			4.22	5.47	77.29	1.83	0.03	5.23
○Thread 27931			4.10	5.47	75.00	1.83	0.03	5.24
○Thread 27932			4.17	5.47	76.28	1.83	0.02	5.23
○Thread 27933			4.17	5.47	76.28	1.83	0.02	5.23
○Thread 27934			4.12	5.47	75.36	1.83	0.02	5.24
○Thread 27935			4.11	5.47	75.26	1.83	0.02	5.23
○Thread 27936			4.26	5.47	77.91	1.83	0.02	5.24
○Thread 27937			4.18	5.47	76.54	1.83	0.02	5.24
○Thread 27938			4.17	5.47	76.26	1.83	0.02	5.22
○Thread 27939			4.11	5.47	75.16	1.83	0.02	5.23
○Thread 27940			4.07	5.47	74.43	1.83	0.02	5.23
○Thread 27941			4.18	5.47	76.43	1.83	0.02	5.23
○Thread 27942			4.08	5.47	74.60	1.83	0.02	5.21
○Thread 27943			4.20	5.47	76.79	1.83	0.02	5.23
○Thread 27944			4.16	5.47	76.15	1.83	0.02	5.23
○Thread 27945			4.09	5.47	74.78	1.83	0.02	5.23
○Thread 27946			4.08	5.47	74.59	1.83	0.02	5.23
○Thread 27947			4.08	5.47	74.68	1.83	0.02	5.23
○Thread 27948			4.24	5.47	77.51	1.83	0.02	5.23
○Thread 27949			4.17	5.47	76.23	1.83	0.02	5.24
○Thread 27950			4.13	5.47	75.58	1.83	0.02	5.23
○Thread 27951			4.11	5.47	75.12	1.83	0.02	5.22
○Thread 27952			4.14	5.47	75.67	1.83	0.02	5.24
○Thread 27953			4.02	5.47	73.56	1.83	0.02	5.23
○Thread 27954			4.17	5.47	76.30	1.83	0.02	5.22
○Thread 27955			4.18	5.47	76.48	1.83	0.02	5.23
○Thread 27956			4.18	5.47	76.39	1.83	0.02	5.23
○Thread 27957			4.10	5.47	74.92	1.83	0.02	5.23
○Thread 27958			4.17	5.47	76.29	1.83	0.02	5.24
○Thread 27959			4.16	5.47	76.01	1.83	0.02	5.22
○Thread 27960			4.20	5.47	76.74	1.83	0.01	5.23
○Thread 27961			4.15	5.47	75.82	1.83	0.01	5.24
○Thread 27962			4.04	5.47	73.81	1.83	0.01	5.23
○Thread 27963			4.17	5.47	76.18	1.83	0.01	5.22
○Thread 27964			4.10	5.47	74.99	1.83	0.01	5.24
○Thread 27965			4.11	5.47	75.08	1.83	0.01	5.23
○Thread 27966			4.09	5.47	74.80	1.83	0.01	5.19
○Thread 27967			4.06	5.47	74.25	1.83	0.01	5.24
○Thread 27968			4.17	5.48	76.16	1.83	0.01	5.23
○Thread 27969			4.17	5.48	76.16	1.83	0.01	5.23
○Thread 27970			4.07	5.48	74.33	1.83	0.01	5.22
○Thread 27971			4.23	5.48	77.25	1.83	0.01	5.23
○Thread 27972			4.12	5.48	75.33	1.83	0.01	5.23
○Thread 27973			4.10	5.48	74.87	1.83	0.01	5.24
○Thread 27974			4.18	5.48	76.33	1.84	0.01	5.24
○Thread 27975			4.11	5.48	75.14	1.84	0.01	5.24
○Thread 27976			4.07	5.48	74.31	1.84	0.01	5.23
○Thread 27977			4.20	5.48	76.77	1.84	0.01	5.24
○Thread 27978			4.14	5.48	75.67	1.84	0.01	5.21
○Thread 27979			4.17	5.48	76.22	1.84	0.01	5.23
○Thread 27980			4.17	5.48	76.12	1.84	0.01	5.24
○Thread 27981			4.03	5.48	73.56	1.84	0.01	5.24
○Thread 27982			4.17	5.48	76.12	1.84	0.01	5.24
○Thread 27983			4.12	5.48	75.29	1.84	0.01	5.24
○Thread 27984			3.98	5.48	72.73	1.84	0.01	5.23
○Thread 27985			4.08	5.48	74.46	1.84	0.01	5.24
○Thread 27986			4.06	5.48	74.09	1.84	0.01	5.23
○Thread 27987			4.26	5.48	77.74	1.84	0.00	5.23
○Thread 27988			4.05	5.48	73.91	1.84	0.00	5.24
○Thread 27989			4.08	5.48	74.45	1.84	0.00	5.23
○Thread 27990			4.11	5.48	75.09	1.84	0.00	5.21
○Thread 27991			4.09	5.48	74.72	1.84	0.00	5.23
○Thread 27992			4.05	5.48	73.98	1.84	0.00	5.24
○Thread 27993			4.10	5.48	74.80	1.84	0.00	5.22
○Thread 27994			4.09	5.48	74.62	1.84	0.00	5.22
○Thread 27995			4.20	5.48	76.62	1.84	0.00	5.20
○Thread 27996			4.22	5.48	76.98	1.84	0.00	5.21
○Thread 27997			3.98	5.48	72.69	1.84	0.00	5.22
○Thread 27998			3.94	5.48	71.87	1.84	0.00	5.23
○Thread 27999			3.84	5.48	70.04	1.84	0.00	5.19

Report Configuration