HPC/Generation-2 nodes/vasp/vasp.lst
< HPC | Generation-2 nodes
Jump to navigation
Jump to search
cores/app=4 Run=03 gen=1 nodes=1 ppn=8 napps=2 tmax=2488.86 s chargecores=8 charge=2.77 core-h/app perf~= 10.0 ops/s/core
cores/app=4 run=01 gen=1 nodes=1 ppn=4 napps=1 tmax=1138.83 s chargecores=8 charge=2.53 core-h/app perf~= 11.0 ops/s/core **1 recommended for charge on gen1
cores/app=4 run=02 gen=2 nodes=1 ppn=4 napps=1 tmax=516.69 s chargecores=8 charge=1.15 core-h/app perf~= 24.2 ops/s/core *
cores/app=4 run=04 gen=2 nodes=1 ppn=8 napps=2 tmax=767.26 s chargecores=8 charge=0.85 core-h/app perf~= 32.6 ops/s/core **2
cores/app=6 run=13 gen=1 nodes=2 ppn=3 napps=1 tmax=816.61 s chargecores=16 charge=3.63 core-h/app perf~= 7.7 ops/s/core
cores/app=6 run=11 gen=1 nodes=1 ppn=6 napps=1 tmax=1566.07 s chargecores=8 charge=3.48 core-h/app perf~= 8.0 ops/s/core
cores/app=6 run=16 gen=1 nodes=2 ppn=6 napps=2 tmax=1401.48 s chargecores=16 charge=3.11 core-h/app perf~= 8.9 ops/s/core
cores/app=6 run=14 gen=2 nodes=2 ppn=3 napps=1 tmax=447.01 s chargecores=16 charge=1.99 core-h/app perf~= 14.0 ops/s/core 5% faster than run=12, but nearly 2x charge
cores/app=6 run=17 gen=2 nodes=2 ppn=6 napps=2 tmax=587.55 s chargecores=16 charge=1.31 core-h/app perf~= 21.3 ops/s/core
cores/app=6 run=12 gen=2 nodes=1 ppn=6 napps=1 tmax=470.10 s chargecores=8 charge=1.04 core-h/app perf~= 26.6 ops/s/core *
cores/app=6 run=15 gen=2 nodes=1 ppn=12 napps=2 tmax=867.12 s chargecores=8 charge=0.96 core-h/app perf~= 28.8 ops/s/core HT
cores/app=8 run=23 gen=1 nodes=2 ppn=4 napps=1 tmax=791.37 s chargecores=16 charge=3.52 core-h/app perf~= 7.9 ops/s/core
cores/app=8 run=26 gen=1 nodes=2 ppn=8 napps=2 tmax=1494.81 s chargecores=16 charge=3.32 core-h/app perf~= 8.4 ops/s/core
cores/app=8 run=21 gen=1 nodes=1 ppn=8 napps=1 tmax=1488.72 s chargecores=8 charge=3.31 core-h/app perf~= 8.4 ops/s/core memory bandwidth congestion
cores/app=8 run=24 gen=2 nodes=2 ppn=4 napps=1 tmax=426.55 s chargecores=16 charge=1.90 core-h/app perf~= 14.7 ops/s/core 10% faster than run=22, but nearly 2x charge
cores/app=8 run=27 gen=2 nodes=2 ppn=8 napps=2 tmax=596.30 s chargecores=16 charge=1.33 core-h/app perf~= 21.0 ops/s/core
cores/app=8 run=22 gen=2 nodes=1 ppn=8 napps=1 tmax=472.16 s chargecores=8 charge=1.05 core-h/app perf~= 26.5 ops/s/core * recommended for charge on gen2
cores/app=8 run=25 gen=2 nodes=1 ppn=16 napps=2 tmax=927.22 s chargecores=8 charge=1.03 core-h/app perf~= 27.0 ops/s/core HT
cores/app=12 run=35 gen=1 nodes=4 ppn=3 napps=1 tmax=503.05 s chargecores=32 charge=4.47 core-h/app perf~= 6.2 ops/s/core
cores/app=12 run=43 gen=1 nodes=4 ppn=6 napps=2 tmax=838.51 s chargecores=32 charge=3.73 core-h/app perf~= 7.5 ops/s/core
cores/app=12 run=31 gen=1 nodes=2 ppn=6 napps=1 tmax=838.08 s chargecores=16 charge=3.72 core-h/app perf~= 7.5 ops/s/core
cores/app=12 run=41 gen=1 nodes=3 ppn=8 napps=2 tmax=1117.20 s chargecores=24 charge=3.72 core-h/app perf~= 7.5 ops/s/core
cores/app=12 run=33 gen=1 nodes=3 ppn=4 napps=1 tmax=544.74 s chargecores=24 charge=3.63 core-h/app perf~= 7.6 ops/s/core recommended for speed on gen1
cores/app=12 run=36 gen=2 nodes=4 ppn=3 napps=1 tmax=267.81 s chargecores=32 charge=2.38 core-h/app perf~= 11.7 ops/s/core
cores/app=12 run=34 gen=2 nodes=3 ppn=4 napps=1 tmax=270.06 s chargecores=24 charge=1.80 core-h/app perf~= 15.4 ops/s/core *
cores/app=12 run=32 gen=2 nodes=2 ppn=6 napps=1 tmax=330.88 s chargecores=16 charge=1.47 core-h/app perf~= 18.9 ops/s/core
cores/app=12 run=44 gen=2 nodes=4 ppn=6 napps=2 tmax=321.16 s chargecores=32 charge=1.43 core-h/app perf~= 19.5 ops/s/core
cores/app=12 run=40 gen=2 nodes=2 ppn=12 napps=2 tmax=582.08 s chargecores=16 charge=1.29 core-h/app perf~= 21.5 ops/s/core HT
cores/app=12 run=42 gen=2 nodes=3 ppn=8 napps=2 tmax=383.44 s chargecores=24 charge=1.28 core-h/app perf~= 21.7 ops/s/core
cores/app=12 run=30 gen=2 nodes=1 ppn=12 napps=1 tmax=565.44 s chargecores=8 charge=1.26 core-h/app perf~= 22.1 ops/s/core HT
cores/app=16 run=56 gen=1 nodes=4 ppn=8 napps=2 tmax=1626.13 s chargecores=32 charge=7.23 core-h/app perf~= 3.8 ops/s/core
cores/app=16 run=53 gen=1 nodes=4 ppn=4 napps=1 tmax=522.50 s chargecores=32 charge=4.64 core-h/app perf~= 6.0 ops/s/core
cores/app=16 run=51 gen=1 nodes=2 ppn=8 napps=1 tmax=998.73 s chargecores=16 charge=4.44 core-h/app perf~= 6.3 ops/s/core ref16 (bad parallel scaling compared to ref8)
cores/app=16 run=54 gen=2 nodes=4 ppn=4 napps=1 tmax=237.59 s chargecores=32 charge=2.11 core-h/app perf~= 13.2 ops/s/core fastest, fairly expensive
cores/app=16 run=52 gen=2 nodes=2 ppn=8 napps=1 tmax=329.10 s chargecores=16 charge=1.46 core-h/app perf~= 19.0 ops/s/core * recommended for speed on gen2
cores/app=16 run=57 gen=2 nodes=4 ppn=8 napps=2 tmax=327.90 s chargecores=32 charge=1.46 core-h/app perf~= 19.1 ops/s/core
cores/app=16 run=55 gen=2 nodes=2 ppn=16 napps=2 tmax=601.78 s chargecores=16 charge=1.34 core-h/app perf~= 20.8 ops/s/core HT
cores/app=16 run=50 gen=2 nodes=1 ppn=16 napps=1 tmax=592.26 s chargecores=8 charge=1.32 core-h/app perf~= 21.1 ops/s/core HT