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 128 -ntg 0 -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/176-131-3962/llama.cpp/run/oneview_runs/multicore/armclang/oneview_run_1761315400OMP_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	126.26
►Process 38521+		1	126.26
○Thread 38521			126.26	127.81	98.79	0.00	0.00	127.79

▶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 128 -ntg 0 -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/176-131-3962/llama.cpp/run/oneview_runs/multicore/armclang/oneview_run_1761315400OMP_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	63.51
►Process 38747+		2	63.51
○Thread 38747			63.51	65.04	97.64	0.00	0.00	63.21
○Thread 38846			62.94	63.22	99.55	1.83	0.00	62.96

▶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 128 -ntg 0 -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/176-131-3962/llama.cpp/run/oneview_runs/multicore/armclang/oneview_run_1761315400OMP_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	32.14
►Process 38915+		4	32.14
○Thread 38915			32.14	33.68	95.44	0.00	0.00	31.84
○Thread 39014			31.54	31.85	99.03	1.83	0.00	31.59
○Thread 39015			31.56	31.85	99.09	1.83	0.00	31.59
○Thread 39016			31.55	31.85	99.06	1.83	0.00	31.59

▶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 128 -ntg 0 -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/176-131-3962/llama.cpp/run/oneview_runs/multicore/armclang/oneview_run_1761315400OMP_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	16.47
►Process 39038+		8	16.47
○Thread 39038			16.47	18.07	91.17	0.00	0.00	16.23
○Thread 39137			15.94	16.23	98.17	1.83	0.00	15.98
○Thread 39138			15.95	16.23	98.23	1.83	0.00	15.98
○Thread 39139			15.94	16.23	98.20	1.83	0.00	15.98
○Thread 39140			15.91	16.23	97.98	1.83	0.00	15.98
○Thread 39141			15.93	16.23	98.13	1.83	0.00	15.98
○Thread 39142			15.94	16.23	98.19	1.83	0.00	15.98
○Thread 39143			15.97	16.23	98.35	1.83	0.00	15.98

▶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 128 -ntg 0 -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/176-131-3962/llama.cpp/run/oneview_runs/multicore/armclang/oneview_run_1761315400OMP_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	8.67
►Process 39213+		16	8.67
○Thread 39213			8.67	10.26	84.46	0.00	0.00	8.43
○Thread 39312			8.13	8.44	96.33	1.81	0.01	8.18
○Thread 39313			8.11	8.44	96.04	1.81	0.01	8.18
○Thread 39314			8.14	8.44	96.39	1.81	0.01	8.18
○Thread 39315			8.12	8.44	96.15	1.81	0.00	8.18
○Thread 39316			8.14	8.44	96.44	1.81	0.00	8.18
○Thread 39317			8.13	8.44	96.32	1.81	0.00	8.18
○Thread 39318			8.12	8.44	96.14	1.81	0.00	8.18
○Thread 39319			8.13	8.44	96.31	1.81	0.00	8.17
○Thread 39320			8.15	8.44	96.49	1.81	0.00	8.18
○Thread 39321			8.15	8.44	96.49	1.81	0.00	8.18
○Thread 39322			8.13	8.44	96.25	1.82	0.00	8.18
○Thread 39323			8.14	8.44	96.42	1.82	0.00	8.18
○Thread 39324			8.15	8.44	96.54	1.82	0.00	8.18
○Thread 39325			8.10	8.44	95.94	1.82	0.00	8.18
○Thread 39326			8.15	8.44	96.53	1.82	0.00	8.18

▶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 128 -ntg 0 -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/176-131-3962/llama.cpp/run/oneview_runs/multicore/armclang/oneview_run_1761315400OMP_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	6.56
►Process 39347+		24	6.56
○Thread 39347			6.56	8.14	80.60	0.00	0.00	6.31
○Thread 39446			6.02	6.32	95.23	1.81	0.01	6.06
○Thread 39447			6.01	6.32	95.07	1.81	0.01	6.06
○Thread 39448			6.02	6.32	95.31	1.81	0.01	6.06
○Thread 39449			6.02	6.32	95.30	1.82	0.01	6.06
○Thread 39450			6.00	6.32	94.91	1.82	0.01	6.06
○Thread 39451			6.00	6.32	94.98	1.82	0.01	6.05
○Thread 39452			6.00	6.32	94.90	1.82	0.01	6.06
○Thread 39453			6.02	6.32	95.29	1.82	0.01	6.06
○Thread 39454			6.01	6.32	95.13	1.82	0.01	6.06
○Thread 39455			6.00	6.32	94.97	1.82	0.01	6.05
○Thread 39456			6.01	6.32	95.05	1.82	0.00	6.06
○Thread 39457			6.01	6.32	95.12	1.82	0.00	6.04
○Thread 39458			6.00	6.32	94.88	1.82	0.00	6.05
○Thread 39459			5.99	6.32	94.80	1.82	0.00	6.06
○Thread 39460			5.99	6.32	94.72	1.82	0.00	6.06
○Thread 39461			6.01	6.32	95.11	1.82	0.00	6.06
○Thread 39462			5.98	6.32	94.55	1.82	0.00	6.06
○Thread 39463			5.99	6.32	94.78	1.82	0.00	6.06
○Thread 39464			6.03	6.32	95.34	1.82	0.00	6.05
○Thread 39465			6.02	6.32	95.17	1.82	0.00	6.05
○Thread 39466			6.03	6.32	95.33	1.82	0.00	6.06
○Thread 39467			6.01	6.32	95.01	1.82	0.00	6.06
○Thread 39468			5.84	6.32	92.39	1.82	0.00	6.06

▶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 128 -ntg 0 -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/176-131-3962/llama.cpp/run/oneview_runs/multicore/armclang/oneview_run_1761315400OMP_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	5.36
►Process 39489+		32	5.36
○Thread 39489			5.36	6.92	77.35	0.00	0.00	5.09
○Thread 39588			4.80	5.10	94.06	1.81	0.01	4.84
○Thread 39589			4.78	5.10	93.67	1.81	0.01	4.84
○Thread 39590			4.80	5.10	94.05	1.81	0.01	4.84
○Thread 39591			4.75	5.10	93.17	1.81	0.01	4.84
○Thread 39592			4.76	5.10	93.26	1.81	0.01	4.84
○Thread 39593			4.76	5.10	93.35	1.81	0.01	4.84
○Thread 39594			4.79	5.10	93.94	1.82	0.01	4.84
○Thread 39595			4.76	5.10	93.25	1.82	0.01	4.83
○Thread 39596			4.76	5.10	93.34	1.82	0.01	4.84
○Thread 39597			4.75	5.10	93.05	1.82	0.01	4.83
○Thread 39598			4.80	5.10	94.02	1.82	0.01	4.84
○Thread 39599			4.78	5.10	93.63	1.82	0.01	4.83
○Thread 39600			4.80	5.10	94.11	1.82	0.01	4.84
○Thread 39601			4.78	5.10	93.72	1.82	0.01	4.84
○Thread 39602			4.80	5.10	94.01	1.82	0.01	4.84
○Thread 39603			4.76	5.10	93.32	1.82	0.01	4.84
○Thread 39604			4.76	5.10	93.31	1.82	0.01	4.84
○Thread 39605			4.78	5.10	93.70	1.82	0.00	4.83
○Thread 39606			4.79	5.10	93.89	1.82	0.00	4.84
○Thread 39607			4.77	5.10	93.40	1.82	0.00	4.83
○Thread 39608			4.76	5.10	93.20	1.82	0.00	4.84
○Thread 39609			4.78	5.10	93.68	1.82	0.00	4.84
○Thread 39610			4.79	5.10	93.78	1.82	0.00	4.84
○Thread 39611			4.77	5.10	93.38	1.82	0.00	4.83
○Thread 39612			4.80	5.10	94.06	1.82	0.00	4.84
○Thread 39613			4.77	5.10	93.37	1.82	0.00	4.84
○Thread 39614			4.78	5.10	93.56	1.82	0.00	4.84
○Thread 39615			4.78	5.10	93.56	1.82	0.00	4.84
○Thread 39616			4.80	5.10	94.05	1.82	0.00	4.84
○Thread 39617			4.81	5.10	94.24	1.82	0.00	4.84
○Thread 39618			4.74	5.10	92.76	1.82	0.00	4.83

▶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 128 -ntg 0 -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/176-131-3962/llama.cpp/run/oneview_runs/multicore/armclang/oneview_run_1761315400OMP_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	4.62
►Process 39687+		40	4.62
○Thread 39687			4.62	6.20	74.47	0.00	0.00	4.35
○Thread 39786			4.06	4.36	93.12	1.82	0.01	4.10
○Thread 39787			4.04	4.36	92.66	1.82	0.01	4.10
○Thread 39788			4.01	4.36	92.08	1.82	0.01	4.10
○Thread 39789			4.06	4.36	93.11	1.82	0.01	4.10
○Thread 39790			4.03	4.36	92.53	1.82	0.01	4.10
○Thread 39791			4.05	4.36	92.87	1.82	0.01	4.10
○Thread 39792			4.04	4.36	92.63	1.82	0.01	4.10
○Thread 39793			4.07	4.36	93.32	1.82	0.01	4.10
○Thread 39794			4.06	4.36	93.20	1.82	0.01	4.10
○Thread 39795			4.05	4.36	92.95	1.82	0.01	4.09
○Thread 39796			4.00	4.36	91.80	1.82	0.01	4.10
○Thread 39797			4.04	4.36	92.71	1.82	0.01	4.09
○Thread 39798			4.04	4.36	92.71	1.82	0.01	4.10
○Thread 39799			4.05	4.36	92.93	1.82	0.01	4.10
○Thread 39800			4.03	4.36	92.47	1.82	0.01	4.09
○Thread 39801			4.04	4.36	92.69	1.82	0.01	4.10
○Thread 39802			4.04	4.36	92.69	1.82	0.01	4.10
○Thread 39803			4.05	4.36	92.91	1.82	0.01	4.10
○Thread 39804			4.02	4.36	92.22	1.82	0.01	4.10
○Thread 39805			4.02	4.36	92.22	1.83	0.01	4.10
○Thread 39806			4.07	4.36	93.24	1.83	0.01	4.10
○Thread 39807			4.00	4.37	91.75	1.83	0.01	4.10
○Thread 39808			4.05	4.37	92.77	1.83	0.01	4.10
○Thread 39809			4.05	4.37	92.88	1.83	0.01	4.10
○Thread 39810			4.04	4.37	92.65	1.83	0.01	4.09
○Thread 39811			4.04	4.37	92.65	1.83	0.01	4.10
○Thread 39812			4.06	4.37	92.98	1.83	0.00	4.10
○Thread 39813			4.02	4.37	92.18	1.83	0.00	4.10
○Thread 39814			4.01	4.37	91.94	1.83	0.00	4.10
○Thread 39815			4.03	4.37	92.40	1.83	0.00	4.10
○Thread 39816			4.04	4.37	92.51	1.83	0.00	4.10
○Thread 39817			4.03	4.37	92.39	1.83	0.00	4.10
○Thread 39818			4.05	4.37	92.84	1.83	0.00	4.10
○Thread 39819			4.01	4.37	91.81	1.83	0.00	4.09
○Thread 39820			4.04	4.37	92.49	1.83	0.00	4.10
○Thread 39821			4.05	4.37	92.71	1.83	0.00	4.10
○Thread 39822			4.02	4.37	92.02	1.83	0.00	4.10
○Thread 39823			4.04	4.37	92.47	1.83	0.00	4.09
○Thread 39824			3.79	4.37	86.75	1.83	0.00	4.10

▶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 128 -ntg 0 -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/176-131-3962/llama.cpp/run/oneview_runs/multicore/armclang/oneview_run_1761315400OMP_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	4.18
►Process 39845+		48	4.18
○Thread 39845			4.18	5.67	73.66	0.00	0.00	3.83
○Thread 39944			3.50	3.85	90.99	1.81	0.02	3.58
○Thread 39945			3.52	3.85	91.50	1.81	0.02	3.59
○Thread 39946			3.54	3.85	92.02	1.81	0.02	3.59
○Thread 39947			3.53	3.85	91.88	1.81	0.02	3.59
○Thread 39948			3.51	3.85	91.23	1.81	0.02	3.58
○Thread 39949			3.55	3.85	92.39	1.81	0.01	3.59
○Thread 39950			3.50	3.85	91.09	1.81	0.01	3.59
○Thread 39951			3.51	3.85	91.21	1.81	0.01	3.58
○Thread 39952			3.52	3.85	91.47	1.81	0.01	3.59
○Thread 39953			3.54	3.85	92.11	1.81	0.01	3.59
○Thread 39954			3.53	3.85	91.72	1.81	0.01	3.59
○Thread 39955			3.51	3.85	91.32	1.81	0.01	3.59
○Thread 39956			3.49	3.85	90.67	1.81	0.01	3.57
○Thread 39957			3.54	3.85	92.09	1.81	0.01	3.59
○Thread 39958			3.51	3.85	91.18	1.81	0.01	3.58
○Thread 39959			3.53	3.85	91.69	1.81	0.01	3.59
○Thread 39960			3.53	3.85	91.68	1.81	0.01	3.58
○Thread 39961			3.55	3.85	92.33	1.81	0.01	3.58
○Thread 39962			3.51	3.85	91.28	1.81	0.01	3.59
○Thread 39963			3.52	3.85	91.41	1.81	0.01	3.59
○Thread 39964			3.51	3.85	91.27	1.81	0.01	3.59
○Thread 39965			3.50	3.85	91.01	1.81	0.01	3.58
○Thread 39966			3.51	3.85	91.13	1.81	0.01	3.59
○Thread 39967			3.52	3.85	91.52	1.81	0.01	3.59
○Thread 39968			3.55	3.85	92.16	1.81	0.01	3.59
○Thread 39969			3.53	3.85	91.76	1.81	0.01	3.59
○Thread 39970			3.50	3.85	90.98	1.81	0.01	3.57
○Thread 39971			3.52	3.85	91.36	1.81	0.01	3.58
○Thread 39972			3.51	3.85	91.23	1.81	0.01	3.58
○Thread 39973			3.53	3.85	91.61	1.81	0.01	3.58
○Thread 39974			3.53	3.85	91.74	1.81	0.01	3.59
○Thread 39975			3.55	3.85	92.12	1.82	0.01	3.59
○Thread 39976			3.50	3.85	90.95	1.82	0.01	3.57
○Thread 39977			3.53	3.85	91.72	1.82	0.01	3.59
○Thread 39978			3.50	3.85	90.94	1.82	0.00	3.59
○Thread 39979			3.54	3.85	91.84	1.82	0.00	3.58
○Thread 39980			3.51	3.85	91.19	1.82	0.00	3.58
○Thread 39981			3.53	3.85	91.57	1.82	0.00	3.59
○Thread 39982			3.52	3.86	91.31	1.82	0.00	3.58
○Thread 39983			3.53	3.86	91.69	1.82	0.00	3.59
○Thread 39984			3.53	3.86	91.55	1.82	0.00	3.58
○Thread 39985			3.51	3.86	91.03	1.82	0.00	3.59
○Thread 39986			3.49	3.86	90.63	1.82	0.00	3.59
○Thread 39987			3.50	3.86	90.89	1.82	0.00	3.59
○Thread 39988			3.52	3.86	91.40	1.82	0.00	3.58
○Thread 39989			3.43	3.86	88.93	1.82	0.00	3.58
○Thread 39990			3.24	3.86	84.13	1.82	0.00	3.59

▶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 128 -ntg 0 -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/176-131-3962/llama.cpp/run/oneview_runs/multicore/armclang/oneview_run_1761315400OMP_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	3.70
►Process 40011+		56	3.70
○Thread 40011			3.70	5.30	69.91	0.00	0.00	3.44
○Thread 40110			3.13	3.46	90.68	1.82	0.02	3.20
○Thread 40111			3.13	3.46	90.53	1.82	0.02	3.20
○Thread 40112			3.16	3.46	91.40	1.82	0.02	3.20
○Thread 40113			3.16	3.46	91.53	1.82	0.02	3.20
○Thread 40114			3.15	3.46	91.09	1.82	0.02	3.20
○Thread 40115			3.14	3.46	90.80	1.82	0.02	3.19
○Thread 40116			3.10	3.46	89.78	1.82	0.02	3.20
○Thread 40117			3.13	3.46	90.50	1.82	0.02	3.20
○Thread 40118			3.13	3.46	90.64	1.82	0.02	3.20
○Thread 40119			3.10	3.46	89.77	1.82	0.02	3.19
○Thread 40120			3.12	3.46	90.34	1.82	0.02	3.19
○Thread 40121			3.12	3.46	90.33	1.82	0.02	3.19
○Thread 40122			3.14	3.46	90.91	1.82	0.02	3.20
○Thread 40123			3.12	3.46	90.18	1.83	0.01	3.20
○Thread 40124			3.12	3.46	90.17	1.83	0.01	3.20
○Thread 40125			3.09	3.46	89.30	1.83	0.01	3.19
○Thread 40126			3.12	3.46	90.30	1.83	0.01	3.19
○Thread 40127			3.13	3.46	90.59	1.83	0.01	3.20
○Thread 40128			3.14	3.46	90.87	1.83	0.01	3.19
○Thread 40129			3.14	3.46	90.86	1.83	0.01	3.20
○Thread 40130			3.13	3.46	90.57	1.83	0.01	3.20
○Thread 40131			3.11	3.46	89.99	1.83	0.01	3.20
○Thread 40132			3.14	3.46	90.85	1.83	0.01	3.20
○Thread 40133			3.11	3.46	89.97	1.83	0.01	3.20
○Thread 40134			3.15	3.46	91.12	1.83	0.01	3.19
○Thread 40135			3.12	3.46	90.25	1.83	0.01	3.20
○Thread 40136			3.08	3.46	88.95	1.83	0.01	3.20
○Thread 40137			3.13	3.46	90.38	1.83	0.01	3.20
○Thread 40138			3.10	3.46	89.66	1.83	0.01	3.20
○Thread 40139			3.13	3.46	90.37	1.83	0.01	3.20
○Thread 40140			3.13	3.46	90.37	1.83	0.01	3.20
○Thread 40141			3.12	3.46	90.22	1.83	0.01	3.20
○Thread 40142			3.12	3.46	90.07	1.83	0.01	3.20
○Thread 40143			3.11	3.46	89.77	1.83	0.01	3.19
○Thread 40144			3.12	3.46	90.05	1.83	0.01	3.19
○Thread 40145			3.10	3.46	89.47	1.83	0.01	3.20
○Thread 40146			3.12	3.47	90.19	1.83	0.01	3.20
○Thread 40147			3.15	3.47	90.90	1.83	0.01	3.19
○Thread 40148			3.11	3.47	89.89	1.83	0.01	3.20
○Thread 40149			3.14	3.47	90.60	1.83	0.01	3.20
○Thread 40150			3.15	3.47	90.88	1.83	0.01	3.20
○Thread 40151			3.12	3.47	90.01	1.83	0.01	3.20
○Thread 40152			3.13	3.47	90.44	1.83	0.00	3.20
○Thread 40153			3.14	3.47	90.58	1.83	0.00	3.19
○Thread 40154			3.11	3.47	89.85	1.83	0.00	3.19
○Thread 40155			3.08	3.47	88.98	1.83	0.00	3.20
○Thread 40156			3.15	3.47	90.85	1.83	0.00	3.20
○Thread 40157			3.12	3.47	89.98	1.83	0.00	3.20
○Thread 40158			3.11	3.47	89.83	1.83	0.00	3.20
○Thread 40159			3.14	3.47	90.69	1.83	0.00	3.19
○Thread 40160			3.14	3.47	90.54	1.83	0.00	3.20
○Thread 40161			3.10	3.47	89.52	1.83	0.00	3.20
○Thread 40162			3.07	3.47	88.65	1.83	0.00	3.20
○Thread 40163			2.87	3.47	82.88	1.83	0.00	3.20
○Thread 40164			2.97	3.47	85.76	1.83	0.00	3.20

▶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 128 -ntg 0 -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/176-131-3962/llama.cpp/run/oneview_runs/multicore/armclang/oneview_run_1761315400OMP_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	3.32
►Process 40185+		64	3.32
○Thread 40185			3.32	4.96	67.02	0.00	0.00	3.10
○Thread 40285			2.78	3.12	89.02	1.82	0.02	2.86
○Thread 40286			2.80	3.12	89.65	1.82	0.02	2.86
○Thread 40287			2.80	3.12	89.65	1.82	0.02	2.86
○Thread 40288			2.77	3.12	88.84	1.82	0.02	2.86
○Thread 40289			2.75	3.12	88.19	1.82	0.02	2.85
○Thread 40290			2.77	3.12	88.67	1.82	0.02	2.85
○Thread 40291			2.75	3.12	88.18	1.82	0.02	2.85
○Thread 40292			2.78	3.12	89.14	1.82	0.02	2.85
○Thread 40293			2.77	3.12	88.81	1.82	0.02	2.85
○Thread 40294			2.77	3.12	88.80	1.82	0.02	2.85
○Thread 40295			2.76	3.13	88.32	1.82	0.02	2.85
○Thread 40296			2.72	3.13	87.19	1.82	0.02	2.85
○Thread 40297			2.77	3.13	88.63	1.82	0.02	2.84
○Thread 40298			2.77	3.13	88.62	1.82	0.02	2.86
○Thread 40299			2.75	3.13	87.97	1.82	0.02	2.84
○Thread 40300			2.76	3.13	88.45	1.82	0.02	2.84
○Thread 40301			2.75	3.13	87.96	1.82	0.02	2.84
○Thread 40302			2.76	3.13	88.43	1.82	0.02	2.84
○Thread 40303			2.75	3.13	87.95	1.82	0.02	2.85
○Thread 40304			2.74	3.13	87.78	1.82	0.02	2.84
○Thread 40305			2.72	3.13	87.14	1.82	0.02	2.85
○Thread 40306			2.74	3.13	87.77	1.82	0.01	2.85
○Thread 40307			2.75	3.13	88.08	1.82	0.01	2.85
○Thread 40308			2.73	3.13	87.28	1.82	0.01	2.85
○Thread 40309			2.77	3.13	88.71	1.82	0.01	2.85
○Thread 40310			2.80	3.13	89.50	1.82	0.01	2.85
○Thread 40311			2.78	3.13	89.02	1.82	0.01	2.83
○Thread 40312			2.78	3.13	88.85	1.82	0.01	2.85
○Thread 40313			2.75	3.13	88.04	1.82	0.01	2.84
○Thread 40314			2.75	3.13	87.88	1.82	0.01	2.85
○Thread 40315			2.75	3.13	88.03	1.82	0.01	2.85
○Thread 40316			2.75	3.13	87.87	1.82	0.01	2.84
○Thread 40317			2.74	3.13	87.70	1.82	0.01	2.85
○Thread 40318			2.77	3.13	88.49	1.82	0.01	2.84
○Thread 40319			2.75	3.13	88.01	1.82	0.01	2.84
○Thread 40320			2.77	3.13	88.64	1.82	0.01	2.85
○Thread 40321			2.72	3.13	87.04	1.82	0.01	2.85
○Thread 40322			2.76	3.13	88.15	1.82	0.01	2.84
○Thread 40323			2.75	3.13	87.82	1.82	0.01	2.84
○Thread 40324			2.76	3.13	88.13	1.82	0.01	2.85
○Thread 40325			2.75	3.13	87.81	1.82	0.01	2.85
○Thread 40326			2.70	3.13	86.37	1.82	0.01	2.85
○Thread 40327			2.76	3.13	88.12	1.82	0.01	2.84
○Thread 40328			2.76	3.13	88.11	1.82	0.01	2.85
○Thread 40329			2.74	3.13	87.46	1.82	0.01	2.85
○Thread 40330			2.74	3.13	87.46	1.82	0.01	2.85
○Thread 40331			2.79	3.13	89.05	1.82	0.01	2.85
○Thread 40332			2.80	3.13	89.52	1.82	0.01	2.85
○Thread 40333			2.76	3.13	88.08	1.82	0.01	2.85
○Thread 40334			2.77	3.13	88.39	1.82	0.01	2.84
○Thread 40335			2.73	3.13	87.27	1.82	0.00	2.85
○Thread 40336			2.75	3.13	87.90	1.82	0.00	2.85
○Thread 40337			2.76	3.13	88.05	1.82	0.00	2.84
○Thread 40338			2.75	3.13	87.89	1.82	0.00	2.85
○Thread 40339			2.74	3.13	87.56	1.82	0.00	2.85
○Thread 40340			2.76	3.14	88.19	1.82	0.00	2.85
○Thread 40341			2.74	3.14	87.39	1.82	0.00	2.85
○Thread 40342			2.76	3.14	88.02	1.82	0.00	2.84
○Thread 40343			2.76	3.14	88.01	1.82	0.00	2.84
○Thread 40344			2.77	3.14	88.32	1.82	0.00	2.85
○Thread 40345			2.75	3.14	87.84	1.82	0.00	2.84
○Thread 40346			2.75	3.14	87.83	1.82	0.00	2.85
○Thread 40347			2.76	3.14	88.14	1.82	0.00	2.84

▶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 128 -ntg 0 -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/176-131-3962/llama.cpp/run/oneview_runs/multicore/armclang/oneview_run_1761315400OMP_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	3.19
►Process 40417+		72	3.19
○Thread 40417			3.19	4.78	66.78	0.00	0.00	2.94
○Thread 40516			2.61	2.95	88.34	1.80	0.03	2.68
○Thread 40517			2.66	2.95	90.02	1.80	0.03	2.69
○Thread 40518			2.63	2.95	89.17	1.80	0.03	2.69
○Thread 40519			2.63	2.96	89.00	1.80	0.02	2.69
○Thread 40520			2.62	2.96	88.82	1.80	0.02	2.69
○Thread 40521			2.59	2.96	87.80	1.80	0.02	2.69
○Thread 40522			2.58	2.96	87.45	1.80	0.02	2.68
○Thread 40523			2.57	2.96	86.94	1.80	0.02	2.69
○Thread 40524			2.63	2.96	88.96	1.80	0.02	2.69
○Thread 40525			2.60	2.96	88.11	1.81	0.02	2.67
○Thread 40526			2.61	2.96	88.44	1.81	0.02	2.68
○Thread 40527			2.62	2.96	88.61	1.81	0.02	2.68
○Thread 40528			2.61	2.96	88.43	1.81	0.02	2.69
○Thread 40529			2.58	2.96	87.24	1.81	0.02	2.69
○Thread 40530			2.60	2.96	87.91	1.81	0.02	2.69
○Thread 40531			2.64	2.96	89.26	1.81	0.02	2.69
○Thread 40532			2.60	2.96	88.07	1.81	0.02	2.68
○Thread 40533			2.59	2.96	87.55	1.81	0.02	2.67
○Thread 40534			2.58	2.96	87.38	1.81	0.02	2.69
○Thread 40535			2.59	2.96	87.71	1.81	0.02	2.69
○Thread 40536			2.60	2.96	88.04	1.81	0.02	2.69
○Thread 40537			2.60	2.96	88.03	1.81	0.02	2.68
○Thread 40538			2.61	2.96	88.36	1.81	0.02	2.68
○Thread 40539			2.61	2.96	88.19	1.81	0.02	2.68
○Thread 40540			2.61	2.96	88.18	1.81	0.02	2.68
○Thread 40541			2.59	2.96	87.67	1.81	0.02	2.67
○Thread 40542			2.59	2.96	87.49	1.81	0.02	2.68
○Thread 40543			2.60	2.96	87.99	1.81	0.02	2.69
○Thread 40544			2.56	2.96	86.46	1.81	0.02	2.68
○Thread 40545			2.58	2.96	87.30	1.81	0.02	2.68
○Thread 40546			2.60	2.96	87.97	1.81	0.02	2.69
○Thread 40547			2.64	2.96	89.31	1.81	0.01	2.68
○Thread 40548			2.60	2.96	87.96	1.81	0.01	2.68
○Thread 40549			2.61	2.96	88.29	1.81	0.01	2.68
○Thread 40550			2.58	2.96	87.27	1.81	0.01	2.69
○Thread 40551			2.59	2.96	87.60	1.81	0.01	2.69
○Thread 40552			2.59	2.96	87.42	1.81	0.01	2.68
○Thread 40553			2.59	2.96	87.42	1.81	0.01	2.67
○Thread 40554			2.56	2.96	86.40	1.81	0.01	2.68
○Thread 40555			2.60	2.96	87.91	1.81	0.01	2.69
○Thread 40556			2.61	2.96	88.24	1.81	0.01	2.69
○Thread 40557			2.58	2.96	87.22	1.81	0.01	2.69
○Thread 40558			2.59	2.96	87.55	1.81	0.01	2.68
○Thread 40559			2.58	2.96	87.04	1.81	0.01	2.67
○Thread 40560			2.60	2.96	87.88	1.81	0.01	2.69
○Thread 40561			2.58	2.96	87.20	1.81	0.01	2.68
○Thread 40562			2.62	2.96	88.37	1.81	0.01	2.69
○Thread 40563			2.60	2.97	87.69	1.81	0.01	2.69
○Thread 40564			2.59	2.97	87.34	1.81	0.01	2.68
○Thread 40565			2.56	2.97	86.49	1.81	0.01	2.68
○Thread 40566			2.60	2.97	87.82	1.81	0.01	2.68
○Thread 40567			2.58	2.97	86.97	1.81	0.01	2.68
○Thread 40568			2.55	2.97	86.12	1.81	0.01	2.68
○Thread 40569			2.60	2.97	87.79	1.81	0.01	2.69
○Thread 40570			2.62	2.97	88.29	1.81	0.01	2.68
○Thread 40571			2.59	2.97	87.44	1.81	0.01	2.68
○Thread 40572			2.61	2.97	88.11	1.81	0.01	2.68
○Thread 40573			2.60	2.97	87.60	1.81	0.01	2.69
○Thread 40574			2.62	2.97	88.43	1.81	0.00	2.69
○Thread 40575			2.61	2.97	87.92	1.81	0.00	2.69
○Thread 40576			2.60	2.97	87.58	1.81	0.00	2.68
○Thread 40577			2.62	2.97	88.41	1.81	0.00	2.69
○Thread 40578			2.56	2.97	86.39	1.81	0.00	2.68
○Thread 40579			2.60	2.97	87.73	1.81	0.00	2.69
○Thread 40580			2.58	2.97	87.05	1.81	0.00	2.68
○Thread 40581			2.63	2.97	88.72	1.81	0.00	2.68
○Thread 40582			2.55	2.97	85.85	1.81	0.00	2.69
○Thread 40583			2.44	2.97	82.31	1.81	0.00	2.69
○Thread 40584			2.42	2.97	81.63	1.81	0.00	2.68
○Thread 40585			2.44	2.97	82.30	1.81	0.00	2.69
○Thread 40586			2.37	2.97	79.94	1.81	0.00	2.68

▶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 128 -ntg 0 -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/176-131-3962/llama.cpp/run/oneview_runs/multicore/armclang/oneview_run_1761315400OMP_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	2.91
►Process 40607+		80	2.91
○Thread 40607			2.91	4.58	63.48	0.00	0.00	2.71
○Thread 40707			2.41	2.74	87.88	1.81	0.03	2.46
○Thread 40708			2.39	2.74	87.14	1.81	0.03	2.47
○Thread 40709			2.39	2.74	87.14	1.81	0.03	2.46
○Thread 40710			2.38	2.74	86.95	1.81	0.03	2.47
○Thread 40711			2.39	2.74	87.12	1.82	0.03	2.47
○Thread 40712			2.39	2.74	87.30	1.82	0.03	2.46
○Thread 40713			2.38	2.74	86.75	1.82	0.02	2.47
○Thread 40714			2.36	2.74	86.19	1.82	0.02	2.47
○Thread 40715			2.37	2.74	86.55	1.82	0.02	2.47
○Thread 40716			2.37	2.74	86.36	1.82	0.02	2.46
○Thread 40717			2.37	2.74	86.36	1.82	0.02	2.46
○Thread 40718			2.36	2.74	86.17	1.82	0.02	2.46
○Thread 40719			2.39	2.74	87.07	1.82	0.02	2.47
○Thread 40720			2.37	2.75	86.34	1.82	0.02	2.47
○Thread 40721			2.39	2.75	87.06	1.82	0.02	2.47
○Thread 40722			2.36	2.75	86.14	1.82	0.02	2.47
○Thread 40723			2.40	2.75	87.41	1.82	0.02	2.47
○Thread 40724			2.37	2.75	86.49	1.82	0.02	2.46
○Thread 40725			2.36	2.75	85.94	1.82	0.02	2.47
○Thread 40726			2.36	2.75	86.12	1.82	0.02	2.48
○Thread 40727			2.36	2.75	86.11	1.82	0.02	2.47
○Thread 40728			2.37	2.75	86.47	1.82	0.02	2.47
○Thread 40729			2.38	2.75	86.64	1.82	0.02	2.46
○Thread 40730			2.37	2.75	86.45	1.82	0.02	2.47
○Thread 40731			2.35	2.75	85.72	1.82	0.02	2.46
○Thread 40732			2.37	2.75	86.26	1.82	0.02	2.46
○Thread 40733			2.39	2.75	86.98	1.82	0.02	2.46
○Thread 40734			2.38	2.75	86.61	1.82	0.02	2.46
○Thread 40735			2.37	2.75	86.42	1.82	0.02	2.47
○Thread 40736			2.39	2.75	87.14	1.82	0.02	2.47
○Thread 40737			2.36	2.75	86.04	1.82	0.02	2.47
○Thread 40738			2.39	2.75	86.94	1.82	0.02	2.45
○Thread 40739			2.37	2.75	86.39	1.82	0.02	2.45
○Thread 40740			2.40	2.75	87.30	1.82	0.02	2.47
○Thread 40741			2.39	2.75	87.11	1.82	0.02	2.46
○Thread 40742			2.41	2.75	87.83	1.82	0.02	2.47
○Thread 40743			2.36	2.75	86.00	1.82	0.01	2.45
○Thread 40744			2.36	2.75	85.81	1.82	0.01	2.45
○Thread 40745			2.37	2.75	86.35	1.82	0.01	2.47
○Thread 40746			2.37	2.75	86.35	1.82	0.01	2.47
○Thread 40747			2.39	2.75	87.07	1.82	0.01	2.47
○Thread 40748			2.38	2.75	86.70	1.82	0.01	2.47
○Thread 40749			2.34	2.75	85.24	1.82	0.01	2.47
○Thread 40750			2.35	2.75	85.59	1.82	0.01	2.47
○Thread 40751			2.36	2.75	85.77	1.82	0.01	2.47
○Thread 40752			2.35	2.75	85.40	1.82	0.01	2.47
○Thread 40753			2.36	2.75	85.76	1.82	0.01	2.47
○Thread 40754			2.36	2.75	85.75	1.82	0.01	2.47
○Thread 40755			2.36	2.75	85.92	1.82	0.01	2.47
○Thread 40756			2.36	2.75	85.92	1.82	0.01	2.46
○Thread 40757			2.36	2.75	85.73	1.82	0.01	2.47
○Thread 40758			2.38	2.75	86.45	1.82	0.01	2.46
○Thread 40759			2.36	2.75	85.72	1.82	0.01	2.47
○Thread 40760			2.39	2.75	86.80	1.82	0.01	2.47
○Thread 40761			2.36	2.75	85.89	1.82	0.01	2.47
○Thread 40762			2.36	2.75	85.70	1.82	0.01	2.47
○Thread 40763			2.36	2.75	85.69	1.82	0.01	2.47
○Thread 40764			2.37	2.75	86.05	1.82	0.01	2.47
○Thread 40765			2.34	2.75	85.13	1.82	0.01	2.47
○Thread 40766			2.36	2.75	85.85	1.82	0.01	2.47
○Thread 40767			2.36	2.75	85.66	1.82	0.01	2.46
○Thread 40768			2.36	2.76	85.84	1.82	0.01	2.47
○Thread 40769			2.36	2.76	85.65	1.82	0.01	2.47
○Thread 40770			2.35	2.76	85.28	1.82	0.01	2.46
○Thread 40771			2.38	2.76	86.36	1.82	0.01	2.46
○Thread 40772			2.34	2.76	84.90	1.82	0.00	2.47
○Thread 40773			2.34	2.76	85.08	1.82	0.00	2.46
○Thread 40774			2.38	2.76	86.34	1.82	0.00	2.47
○Thread 40775			2.36	2.76	85.79	1.82	0.00	2.47
○Thread 40776			2.39	2.76	86.69	1.82	0.00	2.47
○Thread 40777			2.35	2.76	85.41	1.82	0.00	2.47
○Thread 40778			2.38	2.76	86.50	1.82	0.00	2.47
○Thread 40779			2.33	2.76	84.68	1.82	0.00	2.46
○Thread 40780			2.33	2.76	84.49	1.82	0.00	2.47
○Thread 40781			2.38	2.76	86.29	1.82	0.00	2.47
○Thread 40782			2.35	2.76	85.38	1.82	0.00	2.45
○Thread 40783			2.33	2.76	84.47	1.82	0.00	2.47
○Thread 40784			2.33	2.76	84.64	1.82	0.00	2.48
○Thread 40785			2.20	2.76	79.92	1.82	0.00	2.47

▶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 128 -ntg 0 -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/176-131-3962/llama.cpp/run/oneview_runs/multicore/armclang/oneview_run_1761315400OMP_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	2.70
►Process 40809+		88	2.70
○Thread 40809			2.70	4.43	60.89	0.00	0.00	2.55
○Thread 40908			2.29	2.60	88.24	1.80	0.03	2.33
○Thread 40909			2.29	2.60	88.05	1.80	0.03	2.33
○Thread 40910			2.26	2.60	86.89	1.80	0.03	2.33
○Thread 40911			2.22	2.60	85.53	1.80	0.03	2.33
○Thread 40912			2.23	2.60	85.91	1.80	0.03	2.33
○Thread 40913			2.25	2.60	86.48	1.80	0.03	2.33
○Thread 40914			2.28	2.60	87.63	1.80	0.03	2.33
○Thread 40915			2.23	2.60	85.89	1.80	0.03	2.33
○Thread 40916			2.23	2.60	85.69	1.80	0.03	2.33
○Thread 40917			2.24	2.60	86.05	1.80	0.03	2.34
○Thread 40918			2.23	2.60	85.65	1.80	0.03	2.32
○Thread 40919			2.23	2.60	85.84	1.80	0.03	2.32
○Thread 40920			2.24	2.60	86.22	1.80	0.03	2.33
○Thread 40921			2.22	2.60	85.44	1.80	0.03	2.33
○Thread 40922			2.22	2.60	85.43	1.80	0.03	2.33
○Thread 40923			2.25	2.60	86.58	1.80	0.03	2.33
○Thread 40924			2.25	2.60	86.38	1.80	0.02	2.33
○Thread 40925			2.21	2.61	84.83	1.80	0.02	2.31
○Thread 40926			2.23	2.61	85.79	1.80	0.02	2.32
○Thread 40927			2.25	2.61	86.55	1.80	0.02	2.33
○Thread 40928			2.21	2.61	84.81	1.80	0.02	2.32
○Thread 40929			2.24	2.61	85.96	1.80	0.02	2.32
○Thread 40930			2.23	2.61	85.57	1.80	0.02	2.33
○Thread 40931			2.20	2.61	84.41	1.81	0.02	2.33
○Thread 40932			2.23	2.61	85.74	1.81	0.02	2.33
○Thread 40933			2.20	2.61	84.39	1.81	0.02	2.31
○Thread 40934			2.23	2.61	85.54	1.81	0.02	2.32
○Thread 40935			2.26	2.61	86.68	1.81	0.02	2.32
○Thread 40936			2.22	2.61	85.14	1.81	0.02	2.33
○Thread 40937			2.25	2.61	86.28	1.81	0.02	2.33
○Thread 40938			2.22	2.61	85.32	1.81	0.02	2.33
○Thread 40939			2.24	2.61	85.88	1.81	0.02	2.31
○Thread 40940			2.23	2.61	85.68	1.81	0.02	2.32
○Thread 40941			2.22	2.61	85.29	1.81	0.02	2.33
○Thread 40942			2.22	2.61	85.29	1.81	0.02	2.33
○Thread 40943			2.25	2.61	86.24	1.81	0.02	2.32
○Thread 40944			2.23	2.61	85.46	1.81	0.02	2.31
○Thread 40945			2.25	2.61	86.41	1.81	0.02	2.33
○Thread 40946			2.24	2.61	85.83	1.81	0.02	2.33
○Thread 40947			2.21	2.61	84.67	1.81	0.02	2.33
○Thread 40948			2.23	2.61	85.63	1.81	0.02	2.33
○Thread 40949			2.23	2.61	85.62	1.81	0.02	2.33
○Thread 40950			2.20	2.61	84.46	1.81	0.02	2.32
○Thread 40951			2.22	2.61	85.03	1.81	0.02	2.33
○Thread 40952			2.17	2.61	83.30	1.81	0.02	2.33
○Thread 40953			2.22	2.61	85.21	1.81	0.01	2.33
○Thread 40954			2.22	2.61	85.20	1.81	0.01	2.33
○Thread 40955			2.22	2.61	85.19	1.81	0.01	2.32
○Thread 40956			2.22	2.61	85.18	1.81	0.01	2.32
○Thread 40957			2.21	2.61	84.60	1.81	0.01	2.31
○Thread 40958			2.23	2.61	85.55	1.81	0.01	2.33
○Thread 40959			2.25	2.61	86.12	1.81	0.01	2.32
○Thread 40960			2.23	2.61	85.54	1.81	0.01	2.33
○Thread 40961			2.21	2.61	84.76	1.81	0.01	2.32
○Thread 40962			2.25	2.61	86.29	1.81	0.01	2.33
○Thread 40963			2.21	2.61	84.75	1.81	0.01	2.32
○Thread 40964			2.21	2.61	84.74	1.81	0.01	2.32
○Thread 40965			2.22	2.61	85.12	1.81	0.01	2.33
○Thread 40966			2.21	2.61	84.73	1.81	0.01	2.33
○Thread 40967			2.24	2.61	85.68	1.81	0.01	2.33
○Thread 40968			2.24	2.61	85.67	1.81	0.01	2.31
○Thread 40969			2.24	2.61	85.85	1.81	0.01	2.33
○Thread 40970			2.20	2.62	84.32	1.81	0.01	2.33
○Thread 40971			2.20	2.62	84.31	1.81	0.01	2.31
○Thread 40972			2.19	2.62	83.73	1.81	0.01	2.31
○Thread 40973			2.26	2.62	86.59	1.81	0.01	2.32
○Thread 40974			2.21	2.62	84.67	1.81	0.01	2.33
○Thread 40975			2.24	2.62	85.62	1.81	0.01	2.33
○Thread 40976			2.18	2.62	83.51	1.81	0.01	2.33
○Thread 40977			2.25	2.62	85.99	1.81	0.01	2.33
○Thread 40978			2.22	2.62	85.02	1.81	0.01	2.33
○Thread 40979			2.20	2.62	84.06	1.81	0.01	2.32
○Thread 40980			2.22	2.62	85.01	1.81	0.01	2.32
○Thread 40981			2.23	2.62	85.38	1.81	0.01	2.33
○Thread 40982			2.18	2.62	83.28	1.81	0.00	2.31
○Thread 40983			2.20	2.62	84.03	1.81	0.00	2.33
○Thread 40984			2.22	2.62	84.79	1.81	0.00	2.33
○Thread 40985			2.24	2.62	85.55	1.81	0.00	2.33
○Thread 40986			2.24	2.62	85.54	1.81	0.00	2.33
○Thread 40987			2.22	2.62	84.77	1.81	0.00	2.32
○Thread 40988			2.22	2.62	84.76	1.81	0.00	2.33
○Thread 40989			2.23	2.62	85.33	1.81	0.00	2.33
○Thread 40990			2.19	2.62	83.60	1.81	0.00	2.33
○Thread 40991			2.20	2.62	83.97	1.81	0.00	2.31
○Thread 40992			2.16	2.62	82.44	1.81	0.00	2.31
○Thread 40993			2.08	2.62	79.57	1.81	0.00	2.31
○Thread 40994			2.05	2.62	78.23	1.81	0.00	2.34

▶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 128 -ntg 0 -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/176-131-3962/llama.cpp/run/oneview_runs/multicore/armclang/oneview_run_1761315400OMP_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	2.66
►Process 41015+		96	2.66
○Thread 41015			2.66	4.38	60.86	0.00	0.00	2.48
○Thread 41114			2.15	2.51	85.66	1.83	0.03	2.23
○Thread 41115			2.16	2.51	86.05	1.83	0.03	2.23
○Thread 41116			2.16	2.51	86.24	1.83	0.03	2.23
○Thread 41117			2.15	2.51	85.64	1.83	0.03	2.23
○Thread 41118			2.14	2.51	85.43	1.83	0.03	2.22
○Thread 41119			2.12	2.51	84.43	1.83	0.03	2.23
○Thread 41120			2.11	2.51	84.22	1.83	0.03	2.23
○Thread 41121			2.16	2.51	86.20	1.84	0.03	2.23
○Thread 41122			2.13	2.51	85.00	1.84	0.03	2.23
○Thread 41123			2.13	2.51	85.00	1.84	0.03	2.24
○Thread 41124			2.15	2.51	85.79	1.84	0.03	2.23
○Thread 41125			2.14	2.51	85.38	1.84	0.03	2.22
○Thread 41126			2.13	2.51	84.98	1.84	0.03	2.23
○Thread 41127			2.16	2.51	85.96	1.84	0.03	2.23
○Thread 41128			2.13	2.51	84.96	1.84	0.03	2.23
○Thread 41129			2.13	2.51	84.95	1.84	0.03	2.23
○Thread 41130			2.11	2.51	83.95	1.84	0.03	2.23
○Thread 41131			2.13	2.51	84.94	1.84	0.03	2.23
○Thread 41132			2.12	2.51	84.33	1.84	0.03	2.22
○Thread 41133			2.13	2.51	84.92	1.84	0.03	2.24
○Thread 41134			2.13	2.51	84.72	1.84	0.03	2.22
○Thread 41135			2.13	2.51	84.71	1.84	0.03	2.23
○Thread 41136			2.12	2.51	84.30	1.84	0.03	2.23
○Thread 41137			2.12	2.51	84.30	1.84	0.03	2.22
○Thread 41138			2.12	2.52	84.29	1.84	0.03	2.23
○Thread 41139			2.16	2.52	85.87	1.84	0.03	2.23
○Thread 41140			2.13	2.52	84.87	1.84	0.03	2.23
○Thread 41141			2.15	2.52	85.66	1.84	0.03	2.23
○Thread 41142			2.11	2.52	83.86	1.84	0.02	2.24
○Thread 41143			2.13	2.52	84.85	1.84	0.02	2.23
○Thread 41144			2.12	2.52	84.44	1.84	0.02	2.23
○Thread 41145			2.09	2.52	83.05	1.84	0.02	2.23
○Thread 41146			2.13	2.52	84.63	1.84	0.02	2.23
○Thread 41147			2.13	2.52	84.82	1.84	0.02	2.23
○Thread 41148			2.09	2.52	83.22	1.84	0.02	2.23
○Thread 41149			2.13	2.52	84.60	1.84	0.02	2.23
○Thread 41150			2.12	2.52	84.20	1.84	0.02	2.24
○Thread 41151			2.12	2.52	84.39	1.84	0.02	2.23
○Thread 41152			2.14	2.52	84.98	1.84	0.02	2.23
○Thread 41153			2.12	2.52	84.38	1.84	0.02	2.21
○Thread 41154			2.12	2.52	84.17	1.84	0.02	2.23
○Thread 41155			2.12	2.52	84.36	1.84	0.02	2.22
○Thread 41156			2.08	2.52	82.57	1.84	0.02	2.23
○Thread 41157			2.16	2.52	85.93	1.84	0.02	2.23
○Thread 41158			2.09	2.52	82.95	1.84	0.02	2.23
○Thread 41159			2.09	2.52	83.14	1.84	0.02	2.22
○Thread 41160			2.12	2.52	84.12	1.84	0.02	2.23
○Thread 41161			2.10	2.52	83.52	1.84	0.02	2.23
○Thread 41162			2.14	2.52	84.90	1.84	0.02	2.23
○Thread 41163			2.09	2.52	82.91	1.84	0.02	2.23
○Thread 41164			2.12	2.52	84.09	1.84	0.02	2.23
○Thread 41165			2.10	2.52	83.29	1.84	0.02	2.24
○Thread 41166			2.13	2.52	84.48	1.84	0.02	2.23
○Thread 41167			2.15	2.52	85.46	1.84	0.02	2.23
○Thread 41168			2.13	2.52	84.46	1.84	0.02	2.23
○Thread 41169			2.13	2.52	84.45	1.84	0.02	2.23
○Thread 41170			2.11	2.52	83.65	1.84	0.02	2.21
○Thread 41171			2.06	2.52	81.86	1.84	0.01	2.23
○Thread 41172			2.12	2.52	84.03	1.84	0.01	2.23
○Thread 41173			2.15	2.52	85.41	1.84	0.01	2.23
○Thread 41174			2.09	2.52	83.03	1.84	0.01	2.23
○Thread 41175			2.14	2.52	84.77	1.84	0.01	2.24
○Thread 41176			2.14	2.52	84.76	1.84	0.01	2.24
○Thread 41177			2.13	2.52	84.36	1.84	0.01	2.23
○Thread 41178			2.11	2.53	83.76	1.84	0.01	2.24
○Thread 41179			2.09	2.53	82.76	1.84	0.01	2.23
○Thread 41180			2.09	2.53	82.75	1.84	0.01	2.23
○Thread 41181			2.12	2.53	83.93	1.84	0.01	2.23
○Thread 41182			2.08	2.53	82.54	1.84	0.01	2.23
○Thread 41183			2.10	2.53	83.13	1.84	0.01	2.23
○Thread 41184			2.10	2.53	83.32	1.84	0.01	2.23
○Thread 41185			2.13	2.53	84.30	1.84	0.01	2.23
○Thread 41186			2.11	2.53	83.70	1.84	0.01	2.22
○Thread 41187			2.12	2.53	84.09	1.84	0.01	2.23
○Thread 41188			2.11	2.53	83.48	1.84	0.01	2.24
○Thread 41189			2.10	2.53	83.08	1.84	0.01	2.23
○Thread 41190			2.12	2.53	83.87	1.84	0.01	2.23
○Thread 41191			2.10	2.53	83.07	1.84	0.01	2.24
○Thread 41192			2.08	2.53	82.47	1.84	0.01	2.23
○Thread 41193			2.11	2.53	83.45	1.84	0.01	2.24
○Thread 41194			2.11	2.53	83.44	1.84	0.01	2.23
○Thread 41195			2.09	2.53	82.84	1.84	0.01	2.24
○Thread 41196			2.07	2.53	81.84	1.84	0.00	2.23
○Thread 41197			2.10	2.53	83.22	1.84	0.00	2.23
○Thread 41198			2.12	2.53	83.80	1.84	0.00	2.23
○Thread 41199			2.08	2.53	82.41	1.85	0.00	2.22
○Thread 41200			2.11	2.53	83.39	1.85	0.00	2.23
○Thread 41201			2.12	2.53	83.98	1.85	0.00	2.23
○Thread 41202			2.09	2.53	82.78	1.85	0.00	2.23
○Thread 41203			2.11	2.53	83.57	1.85	0.00	2.24
○Thread 41204			2.12	2.53	83.95	1.85	0.00	2.24
○Thread 41205			2.08	2.53	82.36	1.85	0.00	2.23
○Thread 41206			2.01	2.53	79.59	1.85	0.00	2.23
○Thread 41207			1.92	2.53	75.83	1.85	0.00	2.23
○Thread 41208			1.86	2.53	73.65	1.85	0.00	2.23

Report Configuration