LAMMPS benchmark

daiqizhi

#LAMMPS
benchmark input:

https://raw.githubusercontent.com/lammps/lammps/develop/bench/POTENTIALS/data.reax
https://raw.githubusercontent.com/lammps/lammps/develop/bench/POTENTIALS/in.reaxc
https://raw.githubusercontent.com/lammps/lammps/develop/bench/POTENTIALS/ffield.reax

cpu info:

Architecture:        x86_64
CPU op-mode(s):      32-bit, 64-bit
Byte Order:          Little Endian
CPU(s):              56
On-line CPU(s) list: 0-55
Thread(s) per core:  1
Core(s) per socket:  14
Socket(s):           4
NUMA node(s):        4
Vendor ID:           GenuineIntel
CPU family:          6
Model:               85
Model name:          Intel(R) Xeon(R) Gold 6132 CPU @ 2.60GHz
Stepping:            4
CPU MHz:             1000.377
BogoMIPS:            5200.00
Virtualization:      VT-x
L1d cache:           32K
L1i cache:           32K
L2 cache:            1024K
L3 cache:            19712K
NUMA node0 CPU(s):   0-13
NUMA node1 CPU(s):   14-27
NUMA node2 CPU(s):   28-41
NUMA node3 CPU(s):   42-55

result:

Performance: 0.007 ns/day, 3309.910 hours/ns, 0.839 timesteps/s
99.8% CPU use with 2 MPI tasks x 1 OpenMP threads

Performance: 0.025 ns/day, 970.401 hours/ns, 2.863 timesteps/s
99.7% CPU use with 8 MPI tasks x 1 OpenMP threads

Performance: 0.043 ns/day, 552.680 hours/ns, 5.026 timesteps/s
99.7% CPU use with 16 MPI tasks x 1 OpenMP threads

Performance: 0.063 ns/day, 383.630 hours/ns, 7.241 timesteps/s
99.7% CPU use with 28 MPI tasks x 1 OpenMP threads

Performance: 0.073 ns/day, 329.549 hours/ns, 8.429 timesteps/s
99.6% CPU use with 32 MPI tasks x 1 OpenMP threads

Performance: 0.090 ns/day, 267.153 hours/ns, 10.398 timesteps/s
99.5% CPU use with 56 MPI tasks x 1 OpenMP threads

daiqizhi

CPU(s):              80
On-line CPU(s) list: 0-79
Thread(s) per core:  1
Core(s) per socket:  20
Socket(s):           4
NUMA node(s):        4
Vendor ID:           GenuineIntel
CPU family:          6
Model:               85
Model name:          Intel(R) Xeon(R) Gold 6248 CPU @ 2.50GHz
Stepping:            7
CPU MHz:             1000.393
CPU max MHz:         2501.0000
CPU min MHz:         1000.0000
BogoMIPS:            5000.00
Virtualization:      VT-x
L1d cache:           32K
L1i cache:           32K
L2 cache:            1024K
L3 cache:            28160K
NUMA node0 CPU(s):   0-19
NUMA node1 CPU(s):   20-39
NUMA node2 CPU(s):   40-59
NUMA node3 CPU(s):   60-79

result:

Performance: 0.016 ns/day, 1497.796 hours/ns, 1.855 timesteps/s
99.7% CPU use with 4 MPI tasks x 1 OpenMP threads

Performance: 0.028 ns/day, 842.723 hours/ns, 3.296 timesteps/s
99.7% CPU use with 8 MPI tasks x 1 OpenMP threads

Performance: 0.049 ns/day, 494.347 hours/ns, 5.619 timesteps/s
99.6% CPU use with 16 MPI tasks x 1 OpenMP threads

Performance: 0.071 ns/day, 338.529 hours/ns, 8.205 timesteps/s
99.6% CPU use with 32 MPI tasks x 1 OpenMP threads

Performance: 0.086 ns/day, 279.302 hours/ns, 9.945 timesteps/s
99.3% CPU use with 56 MPI tasks x 1 OpenMP threads

Performance: 0.090 ns/day, 266.020 hours/ns, 10.442 timesteps/s
99.3% CPU use with 64 MPI tasks x 1 OpenMP threads

Performance: 0.071 ns/day, 338.480 hours/ns, 8.207 timesteps/s
99.4% CPU use with 32 MPI tasks x 2 OpenMP threads

Performance: 0.089 ns/day, 270.038 hours/ns, 10.287 timesteps/s
99.2% CPU use with 72 MPI tasks x 1 OpenMP threads

Performance: 0.082 ns/day, 292.467 hours/ns, 9.498 timesteps/s
99.0% CPU use with 80 MPI tasks x 1 OpenMP threads

Performance: 0.083 ns/day, 288.669 hours/ns, 9.623 timesteps/s
99.4% CPU use with 40 MPI tasks x 2 OpenMP threads

Performance: 0.056 ns/day, 429.530 hours/ns, 6.467 timesteps/s
99.5% CPU use with 20 MPI tasks x 4 OpenMP threads

daiqizhi

Architecture:        x86_64
CPU op-mode(s):      32-bit, 64-bit
Byte Order:          Little Endian
CPU(s):              48
On-line CPU(s) list: 0-47
Thread(s) per core:  1
Core(s) per socket:  24
Socket(s):           2
NUMA node(s):        2
Vendor ID:           GenuineIntel
CPU family:          6
Model:               85
Model name:          Montage Jintide(R) C6248R
Stepping:            7
CPU MHz:             1199.749
CPU max MHz:         3001.0000
CPU min MHz:         1200.0000
BogoMIPS:            6000.00
Virtualization:      VT-x
L1d cache:           32K
L1i cache:           32K
L2 cache:            1024K
L3 cache:            36608K
NUMA node0 CPU(s):   0-23
NUMA node1 CPU(s):   24-47

result:

Performance: 0.014 ns/day, 1661.709 hours/ns, 1.672 timesteps/s
99.7% CPU use with 4 MPI tasks x 1 OpenMP threads

Performance: 0.027 ns/day, 899.203 hours/ns, 3.089 timesteps/s
99.7% CPU use with 8 MPI tasks x 1 OpenMP threads

Performance: 0.046 ns/day, 516.795 hours/ns, 5.375 timesteps/s
99.7% CPU use with 16 MPI tasks x 1 OpenMP threads

Performance: 0.073 ns/day, 329.998 hours/ns, 8.418 timesteps/s
99.6% CPU use with 32 MPI tasks x 1 OpenMP threads

Performance: 0.080 ns/day, 301.275 hours/ns, 9.220 timesteps/s
99.5% CPU use with 40 MPI tasks x 1 OpenMP threads

Performance: 0.082 ns/day, 292.953 hours/ns, 9.482 timesteps/s
99.2% CPU use with 48 MPI tasks x 1 OpenMP threads

daiqizhi

Architecture:          x86_64
CPU op-mode(s):        32-bit, 64-bit
Byte Order:            Little Endian
CPU(s):                64
On-line CPU(s) list:   0-63
Thread(s) per core:    1
Core(s) per socket:    32
Socket(s):             2
NUMA node(s):          8
Vendor ID:             HygonGenuine
CPU family:            24
Model:                 1
Model name:            Hygon C86 7285H 32-core Processor
Stepping:              1
CPU MHz:               2500.000
CPU max MHz:           2500.0000
CPU min MHz:           1200.0000
BogoMIPS:              5000.25
Virtualization:        AMD-V
L1d cache:             32K
L1i cache:             64K
L2 cache:              512K
L3 cache:              8192K
NUMA node0 CPU(s):     0-7
NUMA node1 CPU(s):     8-15
NUMA node2 CPU(s):     16-23
NUMA node3 CPU(s):     24-31
NUMA node4 CPU(s):     32-39
NUMA node5 CPU(s):     40-47
NUMA node6 CPU(s):     48-55
NUMA node7 CPU(s):     56-63

result:

Performance: 0.006 ns/day, 4090.498 hours/ns, 0.679 timesteps/s
99.9% CPU use with 4 MPI tasks x 1 OpenMP threads

Performance: 0.011 ns/day, 2178.607 hours/ns, 1.275 timesteps/s
99.9% CPU use with 8 MPI tasks x 1 OpenMP threads

Performance: 0.020 ns/day, 1204.457 hours/ns, 2.306 timesteps/s
100.0% CPU use with 16 MPI tasks x 1 OpenMP threads

Performance: 0.034 ns/day, 714.420 hours/ns, 3.888 timesteps/s
99.9% CPU use with 32 MPI tasks x 1 OpenMP threads

Performance: 0.052 ns/day, 462.133 hours/ns, 6.011 timesteps/s
99.8% CPU use with 64 MPI tasks x 1 OpenMP threads

Performance: 0.034 ns/day, 713.651 hours/ns, 3.892 timesteps/s
99.9% CPU use with 32 MPI tasks x 2 OpenMP threads

daiqizhi

Architecture:        x86_64
CPU op-mode(s):      32-bit, 64-bit
Byte Order:          Little Endian
CPU(s):              48
On-line CPU(s) list: 0-47
Thread(s) per core:  1
Core(s) per socket:  24
Socket(s):           2
NUMA node(s):        2
Vendor ID:           GenuineIntel
CPU family:          6
Model:               85
Model name:          Intel(R) Xeon(R) Gold 6240R CPU @ 2.40GHz
Stepping:            7
CPU MHz:             1000.262
CPU max MHz:         2401.0000
CPU min MHz:         1000.0000
BogoMIPS:            4800.00
Virtualization:      VT-x
L1d cache:           32K
L1i cache:           32K
L2 cache:            1024K
L3 cache:            36608K
NUMA node0 CPU(s):   0-23
NUMA node1 CPU(s):   24-47

result:

Performance: 0.013 ns/day, 1800.429 hours/ns, 1.543 timesteps/s
100.0% CPU use with 4 MPI tasks x 1 OpenMP threads

Performance: 0.025 ns/day, 967.028 hours/ns, 2.872 timesteps/s
100.0% CPU use with 8 MPI tasks x 1 OpenMP threads

Performance: 0.044 ns/day, 546.964 hours/ns, 5.079 timesteps/s
99.9% CPU use with 16 MPI tasks x 1 OpenMP threads

Performance: 0.073 ns/day, 330.277 hours/ns, 8.410 timesteps/s
99.9% CPU use with 32 MPI tasks x 1 OpenMP threads

Performance: 0.088 ns/day, 271.836 hours/ns, 10.219 timesteps/s
99.6% CPU use with 48 MPI tasks x 1 OpenMP threads

daiqizhi

Kunpeng 920-4826

make yes-most
make mpi -j 96

Architecture:        aarch64
Byte Order:          Little Endian
CPU(s):              96
On-line CPU(s) list: 0-95
Thread(s) per core:  1
Core(s) per socket:  48
Socket(s):           2
NUMA node(s):        2
Vendor ID:           0x48
Model:               0
Stepping:            0x1
CPU max MHz:         2600.0000
CPU min MHz:         200.0000
BogoMIPS:            200.00
L1d cache:           64K
L1i cache:           64K
L2 cache:            512K
L3 cache:            49152K
NUMA node0 CPU(s):   0-47
NUMA node1 CPU(s):   48-95
Flags:               fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma dcpop asimddp asimdfhm

result:

Performance: 0.015 ns/day, 1632.035 hours/ns, 1.702 timesteps/s
99.7% CPU use with 16 MPI tasks x no OpenMP threads

Performance: 0.025 ns/day, 959.617 hours/ns, 2.895 timesteps/s
99.4% CPU use with 32 MPI tasks x no OpenMP threads

Performance: 0.043 ns/day, 552.122 hours/ns, 5.031 timesteps/s
99.1% CPU use with 72 MPI tasks x no OpenMP threads

Performance: 0.048 ns/day, 498.339 hours/ns, 5.574 timesteps/s
98.9% CPU use with 96 MPI tasks x no OpenMP threads