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-172-31-18-66/176-138-1719/llama.cpp/run/oneview_runs/multicore/armclang/oneview_run_1761384116OMP_PROC_BIND: spreadOMP_DISPLAY_AFFINITY: TRUEOMP_AFFINITY_FORMAT: 'OMP: pid %P tid %i thread %n bound to OS proc set {%A}'OMP_DISPLAY_ENV: TRUEOMP_PLACES: threadsOMP_NUM_THREADS: 1

ID	Observed Processes	Observed Threads	Time(s)	Elapsed Time (s)	Active Time (%)	Start (after process) (s)	End (before process) (s)	Maximum Time on the Same CPU (s)
►Node ip-172-31-18-66+	1	1	212.54
►Process 17300+		1	212.54
○Thread 17300			212.54	214.90	98.91	0.00	0.00	146.41

▶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-172-31-18-66/176-138-1719/llama.cpp/run/oneview_runs/multicore/armclang/oneview_run_1761384116OMP_NUM_THREADS: 2OMP_PROC_BIND: spreadOMP_DISPLAY_AFFINITY: TRUEOMP_AFFINITY_FORMAT: 'OMP: pid %P tid %i thread %n bound to OS proc set {%A}'OMP_DISPLAY_ENV: TRUEOMP_PLACES: threads

ID	Observed Processes	Observed Threads	Time(s)	Elapsed Time (s)	Active Time (%)	Start (after process) (s)	End (before process) (s)	Maximum Time on the Same CPU (s)
►Node ip-172-31-18-66+	1	2	107.22
►Process 17396+		2	107.22
○Thread 17396			107.22	109.57	97.85	0.00	0.00	106.91
○Thread 17463			106.37	106.92	99.48	2.64	0.01	106.50

▶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-172-31-18-66/176-138-1719/llama.cpp/run/oneview_runs/multicore/armclang/oneview_run_1761384116OMP_NUM_THREADS: 4OMP_PROC_BIND: spreadOMP_DISPLAY_AFFINITY: TRUEOMP_AFFINITY_FORMAT: 'OMP: pid %P tid %i thread %n bound to OS proc set {%A}'OMP_DISPLAY_ENV: TRUEOMP_PLACES: threads

ID	Observed Processes	Observed Threads	Time(s)	Elapsed Time (s)	Active Time (%)	Start (after process) (s)	End (before process) (s)	Maximum Time on the Same CPU (s)
►Node ip-172-31-18-66+	1	4	54.42
►Process 17493+		4	54.42
○Thread 17493			54.42	56.68	96.01	0.00	0.00	54.06
○Thread 17560			53.47	54.07	98.89	2.61	0.01	53.64
○Thread 17561			53.48	54.07	98.90	2.61	0.01	53.65
○Thread 17562			53.54	54.07	99.02	2.61	0.01	53.65

▶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-172-31-18-66/176-138-1719/llama.cpp/run/oneview_runs/multicore/armclang/oneview_run_1761384116OMP_NUM_THREADS: 8OMP_PROC_BIND: spreadOMP_DISPLAY_AFFINITY: TRUEOMP_AFFINITY_FORMAT: 'OMP: pid %P tid %i thread %n bound to OS proc set {%A}'OMP_DISPLAY_ENV: TRUEOMP_PLACES: threads

ID	Observed Processes	Observed Threads	Time(s)	Elapsed Time (s)	Active Time (%)	Start (after process) (s)	End (before process) (s)	Maximum Time on the Same CPU (s)
►Node ip-172-31-18-66+	1	8	28.21
►Process 17590+		8	28.21
○Thread 17590			28.21	30.60	92.19	0.00	0.00	27.95
○Thread 17657			27.40	27.96	98.02	2.63	0.01	27.54
○Thread 17658			27.32	27.96	97.74	2.63	0.01	27.53
○Thread 17659			27.29	27.96	97.61	2.63	0.01	27.53
○Thread 17660			27.26	27.96	97.50	2.63	0.01	27.54
○Thread 17661			27.30	27.96	97.67	2.63	0.01	27.53
○Thread 17662			27.35	27.96	97.83	2.63	0.01	27.54
○Thread 17663			27.32	27.96	97.74	2.63	0.01	27.54

▶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-172-31-18-66/176-138-1719/llama.cpp/run/oneview_runs/multicore/armclang/oneview_run_1761384116OMP_NUM_THREADS: 16OMP_PROC_BIND: spreadOMP_DISPLAY_AFFINITY: TRUEOMP_AFFINITY_FORMAT: 'OMP: pid %P tid %i thread %n bound to OS proc set {%A}'OMP_DISPLAY_ENV: TRUEOMP_PLACES: threads

ID	Observed Processes	Observed Threads	Time(s)	Elapsed Time (s)	Active Time (%)	Start (after process) (s)	End (before process) (s)	Maximum Time on the Same CPU (s)
►Node ip-172-31-18-66+	1	16	15.26
►Process 17794+		16	15.26
○Thread 17794			15.26	17.70	86.22	0.00	0.00	15.02
○Thread 17861			14.48	15.03	96.30	2.66	0.01	14.61
○Thread 17862			14.34	15.03	95.44	2.66	0.01	14.60
○Thread 17863			14.38	15.03	95.64	2.66	0.01	14.61
○Thread 17864			14.33	15.03	95.30	2.66	0.01	14.60
○Thread 17865			14.25	15.03	94.81	2.66	0.01	14.61
○Thread 17866			14.27	15.03	94.91	2.66	0.01	14.60
○Thread 17867			14.17	15.03	94.24	2.66	0.01	14.60
○Thread 17868			14.25	15.03	94.81	2.66	0.01	14.61
○Thread 17869			14.26	15.03	94.88	2.66	0.01	14.60
○Thread 17870			14.32	15.03	95.28	2.66	0.01	14.61
○Thread 17871			14.25	15.03	94.81	2.66	0.01	14.61
○Thread 17872			14.31	15.03	95.18	2.66	0.01	14.61
○Thread 17873			14.32	15.03	95.25	2.66	0.01	14.59
○Thread 17874			14.23	15.03	94.65	2.66	0.01	14.60
○Thread 17875			14.21	15.03	94.55	2.66	0.01	14.60

▶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-172-31-18-66/176-138-1719/llama.cpp/run/oneview_runs/multicore/armclang/oneview_run_1761384116OMP_NUM_THREADS: 24OMP_PROC_BIND: spreadOMP_DISPLAY_AFFINITY: TRUEOMP_AFFINITY_FORMAT: 'OMP: pid %P tid %i thread %n bound to OS proc set {%A}'OMP_DISPLAY_ENV: TRUEOMP_PLACES: threads

ID	Observed Processes	Observed Threads	Time(s)	Elapsed Time (s)	Active Time (%)	Start (after process) (s)	End (before process) (s)	Maximum Time on the Same CPU (s)
►Node ip-172-31-18-66+	1	24	11.95
►Process 17902+		24	11.95
○Thread 17902			11.95	14.36	83.18	0.00	0.00	11.67
○Thread 17969			11.02	11.68	94.32	2.67	0.01	11.25
○Thread 17970			11.03	11.68	94.41	2.67	0.01	11.67
○Thread 17971			10.98	11.68	93.98	2.67	0.01	11.25
○Thread 17972			10.95	11.68	93.77	2.67	0.01	11.25
○Thread 17973			10.81	11.68	92.61	2.67	0.01	11.24
○Thread 17974			10.90	11.68	93.38	2.67	0.01	11.25
○Thread 17975			10.83	11.68	92.74	2.67	0.01	11.25
○Thread 17976			10.85	11.68	92.91	2.67	0.01	11.25
○Thread 17977			10.79	11.68	92.40	2.67	0.01	11.25
○Thread 17978			10.86	11.68	92.96	2.68	0.01	11.25
○Thread 17979			10.88	11.68	93.13	2.68	0.01	11.25
○Thread 17980			10.90	11.68	93.35	2.68	0.01	11.25
○Thread 17981			10.85	11.68	92.92	2.68	0.01	11.25
○Thread 17982			10.83	11.68	92.70	2.68	0.01	11.26
○Thread 17983			10.84	11.68	92.79	2.68	0.01	11.25
○Thread 17984			10.85	11.68	92.92	2.68	0.01	11.25
○Thread 17985			10.77	11.68	92.24	2.68	0.01	11.25
○Thread 17986			10.76	11.68	92.19	2.68	0.01	11.24
○Thread 17987			10.83	11.68	92.79	2.68	0.01	11.25
○Thread 17988			10.84	11.68	92.88	2.68	0.01	11.24
○Thread 17989			10.70	11.68	91.68	2.68	0.01	11.25
○Thread 17990			10.78	11.68	92.32	2.68	0.01	11.25
○Thread 17991			10.47	11.68	89.63	2.68	0.01	11.25

▶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-172-31-18-66/176-138-1719/llama.cpp/run/oneview_runs/multicore/armclang/oneview_run_1761384116OMP_NUM_THREADS: 32OMP_PROC_BIND: spreadOMP_DISPLAY_AFFINITY: TRUEOMP_AFFINITY_FORMAT: 'OMP: pid %P tid %i thread %n bound to OS proc set {%A}'OMP_DISPLAY_ENV: TRUEOMP_PLACES: threads

ID	Observed Processes	Observed Threads	Time(s)	Elapsed Time (s)	Active Time (%)	Start (after process) (s)	End (before process) (s)	Maximum Time on the Same CPU (s)
►Node ip-172-31-18-66+	1	32	9.84
►Process 18018+		32	9.84
○Thread 18018			9.84	12.44	79.07	0.00	0.00	9.74
○Thread 18085			9.02	9.75	92.52	2.68	0.01	9.32
○Thread 18086			8.99	9.75	92.27	2.68	0.01	9.31
○Thread 18087			9.04	9.75	92.73	2.68	0.01	9.31
○Thread 18088			8.91	9.75	91.39	2.68	0.01	9.32
○Thread 18089			8.89	9.75	91.19	2.68	0.01	9.30
○Thread 18090			8.86	9.75	90.83	2.68	0.01	9.32
○Thread 18091			8.77	9.75	89.91	2.68	0.01	9.32
○Thread 18092			8.82	9.75	90.48	2.68	0.01	9.32
○Thread 18093			8.88	9.75	91.09	2.68	0.01	9.31
○Thread 18094			8.83	9.75	90.53	2.68	0.01	9.31
○Thread 18095			8.93	9.75	91.66	2.68	0.01	9.32
○Thread 18096			8.78	9.75	90.02	2.68	0.01	9.32
○Thread 18097			8.77	9.75	89.97	2.68	0.01	9.31
○Thread 18098			8.92	9.75	91.45	2.68	0.01	9.31
○Thread 18099			8.75	9.75	89.76	2.68	0.01	9.32
○Thread 18100			8.73	9.75	89.56	2.68	0.01	9.31
○Thread 18101			8.81	9.75	90.33	2.68	0.01	9.32
○Thread 18102			8.85	9.75	90.74	2.68	0.01	9.32
○Thread 18103			8.61	9.75	88.38	2.68	0.01	9.32
○Thread 18104			8.69	9.75	89.15	2.68	0.01	9.32
○Thread 18105			8.72	9.75	89.41	2.68	0.01	9.32
○Thread 18106			8.69	9.75	89.10	2.68	0.01	9.31
○Thread 18107			8.68	9.75	89.10	2.68	0.01	9.32
○Thread 18108			8.74	9.75	89.67	2.68	0.01	9.32
○Thread 18109			8.73	9.75	89.62	2.68	0.01	9.31
○Thread 18110			8.74	9.75	89.67	2.68	0.01	9.31
○Thread 18111			8.67	9.75	88.95	2.68	0.01	9.32
○Thread 18112			8.73	9.75	89.57	2.68	0.01	9.31
○Thread 18113			8.79	9.75	90.18	2.68	0.01	9.31
○Thread 18114			8.71	9.75	89.36	2.68	0.01	9.32
○Thread 18115			8.73	9.75	89.52	2.68	0.01	9.32

▶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-172-31-18-66/176-138-1719/llama.cpp/run/oneview_runs/multicore/armclang/oneview_run_1761384116OMP_NUM_THREADS: 40OMP_PROC_BIND: spreadOMP_DISPLAY_AFFINITY: TRUEOMP_AFFINITY_FORMAT: 'OMP: pid %P tid %i thread %n bound to OS proc set {%A}'OMP_DISPLAY_ENV: TRUEOMP_PLACES: threads

ID	Observed Processes	Observed Threads	Time(s)	Elapsed Time (s)	Active Time (%)	Start (after process) (s)	End (before process) (s)	Maximum Time on the Same CPU (s)
►Node ip-172-31-18-66+	1	40	8.69
►Process 18142+		40	8.69
○Thread 18142			8.69	11.37	76.44	0.00	0.00	8.69
○Thread 18209			7.90	8.70	90.75	2.66	0.01	8.28
○Thread 18210			7.89	8.70	90.75	2.66	0.01	8.28
○Thread 18211			7.86	8.70	90.29	2.66	0.01	8.28
○Thread 18212			7.66	8.70	88.10	2.66	0.01	8.28
○Thread 18213			7.72	8.70	88.68	2.66	0.01	8.28
○Thread 18214			7.63	8.70	87.65	2.66	0.01	8.28
○Thread 18215			7.69	8.70	88.34	2.66	0.01	8.26
○Thread 18216			7.72	8.70	88.68	2.66	0.01	8.28
○Thread 18217			7.62	8.70	87.53	2.66	0.01	8.28
○Thread 18218			7.66	8.70	87.99	2.66	0.01	8.28
○Thread 18219			7.67	8.70	88.11	2.66	0.01	8.28
○Thread 18220			7.63	8.70	87.71	2.66	0.01	8.28
○Thread 18221			7.69	8.70	88.34	2.66	0.01	8.27
○Thread 18222			7.77	8.70	89.26	2.66	0.01	8.28
○Thread 18223			7.67	8.70	88.17	2.66	0.01	8.27
○Thread 18224			7.75	8.70	89.15	2.66	0.01	8.28
○Thread 18225			7.64	8.70	87.83	2.66	0.01	8.28
○Thread 18226			7.70	8.70	88.57	2.66	0.01	8.27
○Thread 18227			7.52	8.70	86.39	2.66	0.01	8.69
○Thread 18228			7.62	8.70	87.66	2.66	0.01	8.28
○Thread 18229			7.71	8.70	88.58	2.66	0.01	8.26
○Thread 18230			7.62	8.70	87.60	2.66	0.01	8.28
○Thread 18231			7.62	8.70	87.66	2.66	0.01	8.28
○Thread 18232			7.67	8.70	88.18	2.66	0.01	8.26
○Thread 18233			7.66	8.70	88.06	2.66	0.01	8.27
○Thread 18234			7.68	8.70	88.23	2.66	0.01	8.27
○Thread 18235			7.68	8.70	88.24	2.66	0.01	8.28
○Thread 18236			7.69	8.70	88.35	2.66	0.01	8.28
○Thread 18237			7.66	8.70	88.12	2.66	0.01	8.28
○Thread 18238			7.62	8.70	87.60	2.66	0.01	8.28
○Thread 18239			7.60	8.70	87.38	2.66	0.01	8.27
○Thread 18240			7.75	8.70	89.10	2.66	0.01	8.28
○Thread 18241			7.68	8.70	88.24	2.66	0.01	8.28
○Thread 18242			7.62	8.70	87.61	2.66	0.01	8.28
○Thread 18243			7.59	8.70	87.32	2.66	0.01	8.28
○Thread 18244			7.67	8.70	88.13	2.66	0.01	8.25
○Thread 18245			7.48	8.70	86.00	2.66	0.01	8.28
○Thread 18246			7.58	8.70	87.09	2.66	0.01	8.27
○Thread 18247			7.10	8.70	81.63	2.66	0.01	8.28

▶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-172-31-18-66/176-138-1719/llama.cpp/run/oneview_runs/multicore/armclang/oneview_run_1761384116OMP_NUM_THREADS: 48OMP_PROC_BIND: spreadOMP_DISPLAY_AFFINITY: TRUEOMP_AFFINITY_FORMAT: 'OMP: pid %P tid %i thread %n bound to OS proc set {%A}'OMP_DISPLAY_ENV: TRUEOMP_PLACES: threads

ID	Observed Processes	Observed Threads	Time(s)	Elapsed Time (s)	Active Time (%)	Start (after process) (s)	End (before process) (s)	Maximum Time on the Same CPU (s)
►Node ip-172-31-18-66+	1	48	7.85
►Process 18274+		48	7.85
○Thread 18274			7.85	10.58	74.12	0.00	0.00	7.92
○Thread 18341			7.08	7.94	89.27	2.64	0.01	7.51
○Thread 18342			7.13	7.94	89.77	2.64	0.01	7.49
○Thread 18343			7.07	7.94	89.14	2.64	0.01	7.51
○Thread 18344			6.98	7.94	87.95	2.64	0.01	7.51
○Thread 18345			6.78	7.94	85.49	2.64	0.01	7.51
○Thread 18346			6.72	7.94	84.67	2.64	0.01	7.51
○Thread 18347			6.82	7.94	85.99	2.64	0.01	7.51
○Thread 18348			6.86	7.94	86.44	2.64	0.01	7.52
○Thread 18349			6.80	7.94	85.74	2.64	0.01	7.51
○Thread 18350			6.87	7.94	86.57	2.64	0.01	7.51
○Thread 18351			6.86	7.94	86.38	2.64	0.01	7.51
○Thread 18352			6.88	7.94	86.69	2.64	0.01	7.51
○Thread 18353			6.84	7.94	86.19	2.64	0.01	7.51
○Thread 18354			6.88	7.94	86.63	2.64	0.01	7.51
○Thread 18355			6.83	7.94	86.13	2.64	0.01	7.52
○Thread 18356			6.87	7.94	86.57	2.64	0.01	7.52
○Thread 18357			6.77	7.94	85.37	2.64	0.01	7.51
○Thread 18358			6.75	7.94	85.06	2.64	0.01	7.51
○Thread 18359			6.86	7.94	86.45	2.64	0.01	7.51
○Thread 18360			6.76	7.94	85.25	2.64	0.01	7.51
○Thread 18361			6.74	7.94	84.94	2.64	0.01	7.50
○Thread 18362			6.74	7.94	84.94	2.64	0.01	7.51
○Thread 18363			6.68	7.93	84.18	2.64	0.01	7.49
○Thread 18364			6.76	7.93	85.19	2.64	0.01	7.50
○Thread 18365			6.76	7.93	85.13	2.64	0.01	7.50
○Thread 18366			6.86	7.93	86.39	2.64	0.01	7.51
○Thread 18367			6.84	7.93	86.14	2.64	0.01	7.51
○Thread 18368			6.79	7.93	85.64	2.64	0.01	7.51
○Thread 18369			6.74	7.93	84.88	2.64	0.01	7.51
○Thread 18370			6.76	7.93	85.13	2.64	0.01	7.51
○Thread 18371			6.81	7.93	85.83	2.64	0.01	7.51
○Thread 18372			6.91	7.93	87.09	2.64	0.01	7.51
○Thread 18373			6.73	7.93	84.88	2.64	0.01	7.51
○Thread 18374			6.82	7.93	85.89	2.64	0.01	7.51
○Thread 18375			6.85	7.93	86.33	2.64	0.01	7.51
○Thread 18376			6.82	7.93	85.89	2.64	0.01	7.51
○Thread 18377			6.70	7.93	84.38	2.64	0.01	7.51
○Thread 18378			6.74	7.93	84.89	2.64	0.01	7.51
○Thread 18379			6.68	7.93	84.19	2.64	0.01	7.51
○Thread 18380			6.90	7.93	86.97	2.64	0.01	7.51
○Thread 18381			6.77	7.93	85.27	2.64	0.01	7.51
○Thread 18382			6.81	7.93	85.83	2.64	0.01	7.51
○Thread 18383			6.76	7.93	85.14	2.64	0.01	7.52
○Thread 18384			6.64	7.93	83.76	2.64	0.01	7.50
○Thread 18385			6.68	7.93	84.13	2.64	0.01	7.51
○Thread 18386			6.59	7.93	83.06	2.64	0.01	7.50
○Thread 18387			6.20	7.93	78.09	2.64	0.01	7.51

▶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-172-31-18-66/176-138-1719/llama.cpp/run/oneview_runs/multicore/armclang/oneview_run_1761384116OMP_NUM_THREADS: 56OMP_PROC_BIND: spreadOMP_DISPLAY_AFFINITY: TRUEOMP_AFFINITY_FORMAT: 'OMP: pid %P tid %i thread %n bound to OS proc set {%A}'OMP_DISPLAY_ENV: TRUEOMP_PLACES: threads

ID	Observed Processes	Observed Threads	Time(s)	Elapsed Time (s)	Active Time (%)	Start (after process) (s)	End (before process) (s)	Maximum Time on the Same CPU (s)
►Node ip-172-31-18-66+	1	56	7.25
►Process 18414+		56	7.25
○Thread 18414			7.25	10.14	71.44	0.00	0.00	7.44
○Thread 18481			6.61	7.45	88.75	2.68	0.01	7.02
○Thread 18482			6.65	7.45	89.29	2.68	0.01	7.03
○Thread 18483			6.61	7.45	88.75	2.68	0.01	7.03
○Thread 18484			6.44	7.45	86.53	2.68	0.01	7.02
○Thread 18485			6.34	7.45	85.06	2.68	0.01	7.02
○Thread 18486			6.20	7.45	83.31	2.68	0.01	7.03
○Thread 18487			6.21	7.45	83.31	2.68	0.01	6.99
○Thread 18488			6.32	7.45	84.86	2.68	0.01	7.01
○Thread 18489			6.27	7.45	84.26	2.68	0.01	7.03
○Thread 18490			6.23	7.45	83.65	2.68	0.01	7.01
○Thread 18491			6.31	7.45	84.66	2.68	0.01	7.02
○Thread 18492			6.29	7.45	84.46	2.68	0.01	7.03
○Thread 18493			6.28	7.45	84.33	2.68	0.01	7.02
○Thread 18494			6.11	7.45	82.11	2.68	0.01	7.03
○Thread 18495			6.20	7.45	83.32	2.68	0.01	7.01
○Thread 18496			6.32	7.45	84.87	2.68	0.01	7.02
○Thread 18497			6.14	7.45	82.45	2.68	0.01	7.01
○Thread 18498			6.11	7.45	82.05	2.68	0.01	7.00
○Thread 18499			6.22	7.45	83.53	2.68	0.01	7.02
○Thread 18500			6.32	7.45	84.81	2.68	0.01	7.02
○Thread 18501			6.16	7.45	82.72	2.68	0.01	7.02
○Thread 18502			6.17	7.45	82.86	2.68	0.01	7.03
○Thread 18503			6.20	7.45	83.20	2.68	0.01	7.02
○Thread 18504			6.21	7.45	83.40	2.69	0.01	7.03
○Thread 18505			6.20	7.45	83.26	2.69	0.01	7.03
○Thread 18506			6.19	7.45	83.20	2.69	0.01	7.02
○Thread 18507			6.15	7.45	82.53	2.69	0.01	7.02
○Thread 18508			6.15	7.45	82.53	2.69	0.01	7.02
○Thread 18509			6.16	7.45	82.73	2.69	0.01	7.02
○Thread 18510			6.05	7.45	81.25	2.69	0.01	7.02
○Thread 18511			6.26	7.45	84.14	2.69	0.01	7.03
○Thread 18512			6.22	7.45	83.54	2.69	0.01	7.02
○Thread 18513			6.18	7.45	83.07	2.69	0.01	7.03
○Thread 18514			6.19	7.45	83.14	2.69	0.01	7.03
○Thread 18515			6.22	7.45	83.47	2.69	0.01	7.03
○Thread 18516			6.25	7.45	84.01	2.69	0.01	7.01
○Thread 18517			6.09	7.45	81.79	2.69	0.01	7.03
○Thread 18518			6.27	7.45	84.28	2.69	0.01	7.03
○Thread 18519			6.22	7.45	83.47	2.69	0.01	7.01
○Thread 18520			6.17	7.45	82.87	2.69	0.01	7.03
○Thread 18521			6.13	7.45	82.33	2.69	0.01	7.03
○Thread 18522			6.24	7.45	83.74	2.69	0.01	7.03
○Thread 18523			6.08	7.45	81.73	2.69	0.01	7.01
○Thread 18524			6.08	7.45	81.66	2.69	0.01	7.03
○Thread 18525			6.14	7.45	82.54	2.69	0.01	7.02
○Thread 18526			6.16	7.45	82.74	2.69	0.01	7.03
○Thread 18527			6.15	7.45	82.67	2.69	0.01	7.02
○Thread 18528			6.25	7.45	83.95	2.69	0.01	7.03
○Thread 18529			6.20	7.44	83.28	2.69	0.01	7.03
○Thread 18530			6.15	7.44	82.61	2.69	0.01	7.02
○Thread 18531			6.14	7.44	82.54	2.69	0.01	7.02
○Thread 18532			6.02	7.44	80.93	2.69	0.01	7.01
○Thread 18533			6.07	7.44	81.60	2.69	0.01	7.02
○Thread 18534			5.71	7.44	76.70	2.69	0.01	7.01
○Thread 18535			5.64	7.44	75.69	2.69	0.01	7.03

▶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-172-31-18-66/176-138-1719/llama.cpp/run/oneview_runs/multicore/armclang/oneview_run_1761384116OMP_NUM_THREADS: 64OMP_PROC_BIND: spreadOMP_DISPLAY_AFFINITY: TRUEOMP_AFFINITY_FORMAT: 'OMP: pid %P tid %i thread %n bound to OS proc set {%A}'OMP_DISPLAY_ENV: TRUEOMP_PLACES: threads

ID	Observed Processes	Observed Threads	Time(s)	Elapsed Time (s)	Active Time (%)	Start (after process) (s)	End (before process) (s)	Maximum Time on the Same CPU (s)
►Node ip-172-31-18-66+	1	64	6.82
►Process 18562+		64	6.82
○Thread 18562			6.82	9.86	69.16	0.00	0.00	7.14
○Thread 18629			6.18	7.16	86.33	2.69	0.01	6.73
○Thread 18630			6.22	7.16	86.96	2.69	0.01	6.73
○Thread 18631			6.27	7.16	87.52	2.69	0.01	6.73
○Thread 18632			6.08	7.16	84.86	2.69	0.01	6.73
○Thread 18633			6.04	7.16	84.38	2.69	0.01	6.73
○Thread 18634			5.90	7.16	82.42	2.69	0.01	6.73
○Thread 18635			5.85	7.16	81.72	2.69	0.01	6.73
○Thread 18636			6.03	7.16	84.24	2.69	0.01	6.74
○Thread 18637			5.80	7.16	81.02	2.69	0.01	6.73
○Thread 18638			5.76	7.16	80.40	2.69	0.01	6.73
○Thread 18639			5.80	7.16	81.03	2.69	0.01	6.73
○Thread 18640			5.91	7.16	82.57	2.69	0.01	6.73
○Thread 18641			5.73	7.16	80.05	2.69	0.01	6.73
○Thread 18642			5.81	7.16	81.17	2.69	0.01	6.73
○Thread 18643			5.80	7.16	81.10	2.69	0.01	6.73
○Thread 18644			5.79	7.16	80.96	2.69	0.01	6.73
○Thread 18645			5.70	7.16	79.63	2.69	0.01	6.73
○Thread 18646			5.75	7.16	80.40	2.69	0.01	6.72
○Thread 18647			5.80	7.16	80.96	2.69	0.01	6.73
○Thread 18648			5.78	7.16	80.75	2.69	0.01	6.73
○Thread 18649			5.74	7.16	80.13	2.69	0.01	6.73
○Thread 18650			5.76	7.16	80.41	2.69	0.01	6.73
○Thread 18651			5.75	7.16	80.34	2.69	0.01	6.73
○Thread 18652			5.70	7.16	79.71	2.69	0.01	6.73
○Thread 18653			5.71	7.16	79.85	2.69	0.01	6.73
○Thread 18654			5.64	7.16	78.87	2.69	0.01	6.74
○Thread 18655			5.74	7.16	80.27	2.69	0.01	6.72
○Thread 18656			5.88	7.16	82.09	2.69	0.01	6.72
○Thread 18657			5.78	7.16	80.83	2.69	0.01	6.73
○Thread 18658			5.68	7.16	79.36	2.69	0.01	6.73
○Thread 18659			5.78	7.16	80.76	2.69	0.01	6.73
○Thread 18660			5.89	7.16	82.30	2.69	0.01	6.73
○Thread 18661			5.76	7.16	80.55	2.69	0.01	6.73
○Thread 18662			5.74	7.16	80.27	2.69	0.01	6.72
○Thread 18663			5.69	7.16	79.44	2.69	0.01	6.73
○Thread 18664			5.92	7.16	82.65	2.69	0.01	6.73
○Thread 18665			5.81	7.16	81.25	2.69	0.01	6.73
○Thread 18666			5.75	7.16	80.42	2.69	0.01	6.73
○Thread 18667			5.61	7.16	78.39	2.69	0.01	6.73
○Thread 18668			5.75	7.16	80.35	2.69	0.01	6.73
○Thread 18669			5.75	7.16	80.35	2.69	0.01	6.72
○Thread 18670			5.81	7.16	81.19	2.69	0.01	6.73
○Thread 18671			5.65	7.16	78.95	2.69	0.01	6.74
○Thread 18672			5.66	7.16	79.16	2.69	0.01	6.73
○Thread 18673			5.76	7.16	80.42	2.69	0.01	6.73
○Thread 18674			5.68	7.16	79.37	2.70	0.01	6.72
○Thread 18675			5.84	7.16	81.68	2.70	0.01	6.73
○Thread 18676			5.71	7.16	79.79	2.70	0.01	6.73
○Thread 18677			5.90	7.16	82.45	2.70	0.01	6.73
○Thread 18678			5.69	7.16	79.51	2.70	0.01	6.72
○Thread 18679			5.60	7.16	78.26	2.70	0.01	6.73
○Thread 18680			5.76	7.16	80.49	2.70	0.01	6.73
○Thread 18681			5.70	7.16	79.66	2.70	0.01	6.73
○Thread 18682			5.85	7.16	81.68	2.70	0.01	6.73
○Thread 18683			5.62	7.16	78.54	2.70	0.01	6.73
○Thread 18684			5.80	7.16	81.05	2.70	0.01	6.73
○Thread 18685			5.75	7.16	80.36	2.70	0.01	6.73
○Thread 18686			5.68	7.16	79.31	2.70	0.01	6.73
○Thread 18687			5.82	7.16	81.41	2.70	0.01	6.74
○Thread 18688			5.76	7.16	80.57	2.70	0.01	6.73
○Thread 18689			5.78	7.16	80.85	2.70	0.01	6.73
○Thread 18690			5.69	7.16	79.59	2.70	0.01	6.73
○Thread 18691			5.78	7.16	80.78	2.70	0.01	6.73

Report Configuration