Commit a4bc233d authored by sjplimp's avatar sjplimp Committed by GitHub
Browse files

Merge pull request #525 from akohlmey/user-tally-refactor 

Refactoring of USER-TALLY computes to handle sparse/hybrid system for many processors plus bugfixes
parents 5c5b4ffa 11751521

examples/USER/tally/log.21Aug15.force.1→examples/USER/tally/log.12Jun17.force.1

+177 −0
Original line number Diff line number Diff line 
LAMMPS (21 Aug 2015-ICMS)

  using 1 OpenMP thread(s) per MPI task

LAMMPS (19 May 2017)



units		real

atom_style	full
@@ -50,10 +49,39 @@ fix 1 all shake 0.0001 20 0 b 1 a 1 
  1500 = # of frozen angles

fix		2 all nvt temp 300.0 300.0 100.0



# make certain that shake constraints are satisfied

run 0 post no

PPPM initialization ...

WARNING: Using 12-bit tables for long-range coulomb (../kspace.cpp:321)

  G vector (1/distance) = 0.218482

  grid = 15 15 15

  stencil order = 5

  estimated absolute RMS force accuracy = 0.0319435

  estimated relative force accuracy = 9.61968e-05

  using double precision FFTs

  3d grid and FFT values/proc = 8000 3375

Neighbor list info ...

  update every 1 steps, delay 10 steps, check yes

  max neighbors/atom: 2000, page size: 100000

  master list distance cutoff = 14

  ghost atom cutoff = 14

  binsize = 7, bins = 6 6 6

  1 neighbor lists, perpetual/occasional/extra = 1 0 0

  (1) pair lj/cut/coul/long, perpetual

      attributes: half, newton on

      pair build: half/bin/newton

      stencil: half/bin/3d/newton

      bin: standard

Per MPI rank memory allocation (min/avg/max) = 26.54 | 26.54 | 26.54 Mbytes

Step Temp E_pair E_mol TotEng Press 

       0            0   -16692.358            0   -16692.358   -1289.8319 

Loop time of 3e-06 on 1 procs for 0 steps with 4500 atoms





group		one molecule 1 2

6 atoms in group one



# the following section shows equivalences between using the pe/tally compute and other computes and thermo keywords

# the following section shows equivalences between using the force/tally compute and other computes and thermo keywords



# compute pairwise force between two molecules and everybody

compute		fpa one group/group all pair yes kspace no boundary no
@@ -69,17 +97,19 @@ variable fpa equal sqrt(c_fpa[1]*c_fpa[1]+c_fpa[2]*c_fpa[2]+c_fpa[3]*c_fpa[3]) 
variable	for equal sqrt(c_one[1]*c_one[1]+c_one[2]*c_one[2]+c_one[3]*c_one[3])

# round to 10**-10 absolute precision.

variable	ref equal round(1e10*sqrt(c_red[1]*c_red[1]+c_red[2]*c_red[2]+c_red[3]*c_red[3]))*1e-10

variable	all equal round(1e10*c_c2)*1e-10



velocity	all create 300 432567 dist uniform



timestep	2.0



# v_fpa and v_for and c_c1, c_fpa[] and c_one[] should all each have the same value. v_ref and c_c2 should be zero

thermo_style    custom step v_fpa v_for c_c1 c_fpa[1] c_one[1] c_fpa[2] c_one[2] c_fpa[3] c_one[3] v_ref  c_c2

thermo_style    custom step v_fpa v_for c_c1 c_fpa[1] c_one[1] c_fpa[2] c_one[2] c_fpa[3] c_one[3] v_ref v_all

thermo		10



run 50

PPPM initialization ...

WARNING: Using 12-bit tables for long-range coulomb (../kspace.cpp:321)

  G vector (1/distance) = 0.218482

  grid = 15 15 15

  stencil order = 5
@@ -87,38 +117,49 @@ PPPM initialization ... 
  estimated relative force accuracy = 9.61968e-05

  using double precision FFTs

  3d grid and FFT values/proc = 8000 3375

WARNING: Compute force/tally only called from pair style (../compute_force_tally.cpp:75)

WARNING: Compute force/tally only called from pair style (../compute_force_tally.cpp:75)

WARNING: Compute force/tally only called from pair style (../compute_force_tally.cpp:77)

WARNING: Compute force/tally only called from pair style (../compute_force_tally.cpp:77)

Neighbor list info ...

  2 neighbor list requests

  update every 1 steps, delay 10 steps, check yes

  max neighbors/atom: 2000, page size: 100000

  master list distance cutoff = 14

  ghost atom cutoff = 14

  binsize = 7 -> bins = 6 6 6

Memory usage per processor = 16.7648 Mbytes

Step fpa for c1 fpa[1] one[1] fpa[2] one[2] fpa[3] one[3] ref c2 

       0    22.732789    22.732789    22.732789   -17.068392   -17.068392   -8.8345214   -8.8345214   -12.140878   -12.140878            0            0 

      10    11.736915    11.736915    11.736915   -3.3898298   -3.3898298     9.119272     9.119272   -6.5652948   -6.5652948            0            0 

      20    5.6119761    5.6119761    5.6119761  -0.60028931  -0.60028931   -4.4479886   -4.4479886     3.368876     3.368876            0            0 

      30    17.292617    17.292617    17.292617    6.1793856    6.1793856   -10.593927   -10.593927    12.190919    12.190919            0            0 

      40    18.664226    18.664226    18.664226    5.4725079    5.4725079    -6.933046    -6.933046    16.441955    16.441955            0            0 

      50    12.130282    12.130282    12.130282   -1.0321244   -1.0321244    8.0032646    8.0032646   -9.0568326   -9.0568326            0            0 



Loop time of 4.11825 on 1 procs for 50 steps with 4500 atoms

100.0% CPU use with 1 MPI tasks x 1 OpenMP threads

Performance: 2.098 ns/day  11.440 hours/ns  12.141 timesteps/s



MPI task timings breakdown:

  binsize = 7, bins = 6 6 6

  2 neighbor lists, perpetual/occasional/extra = 1 1 0

  (1) pair lj/cut/coul/long, perpetual

      attributes: half, newton on

      pair build: half/bin/newton

      stencil: half/bin/3d/newton

      bin: standard

  (2) compute group/group, occasional, copy from (1)

      attributes: half, newton on

      pair build: copy

      stencil: none

      bin: none

Per MPI rank memory allocation (min/avg/max) = 28.47 | 28.47 | 28.47 Mbytes

Step v_fpa v_for c_c1 c_fpa[1] c_one[1] c_fpa[2] c_one[2] c_fpa[3] c_one[3] v_ref v_all 

       0      22.7331      22.7331      22.7331   -17.068295   -17.068295   -8.8348335   -8.8348334   -12.141369   -12.141369            0            0 

      10    11.736901    11.736901    11.736901   -3.3897029   -3.3897029    9.1193856    9.1193856   -6.5651786   -6.5651786            0            0 

      20    5.6120339    5.6120339    5.6120339  -0.60046861  -0.60046861   -4.4481306   -4.4481306    3.3687528    3.3687528            0            0 

      30     17.29261     17.29261     17.29261     6.179302     6.179302   -10.593979   -10.593979    12.190906    12.190906            0            0 

      40    18.664433    18.664433    18.664433    5.4727782    5.4727782   -6.9329319   -6.9329319    16.442148    16.442148            0            0 

      50    12.130407    12.130407    12.130407   -1.0321196   -1.0321196    8.0035558    8.0035558   -9.0567428   -9.0567428            0            0 

Loop time of 13.9507 on 1 procs for 50 steps with 4500 atoms



Performance: 0.619 ns/day, 38.752 hours/ns, 3.584 timesteps/s

32.0% CPU use with 1 MPI tasks x no OpenMP threads



MPI task timing breakdown:

Section |  min time  |  avg time  |  max time  |%varavg| %total

---------------------------------------------------------------

Pair    | 3.5286     | 3.5286     | 3.5286     |   0.0 | 85.68

Bond    | 6.1274e-05 | 6.1274e-05 | 6.1274e-05 |   0.0 |  0.00

Kspace  | 0.1937     | 0.1937     | 0.1937     |   0.0 |  4.70

Neigh   | 0.31454    | 0.31454    | 0.31454    |   0.0 |  7.64

Comm    | 0.01037    | 0.01037    | 0.01037    |   0.0 |  0.25

Output  | 0.039355   | 0.039355   | 0.039355   |   0.0 |  0.96

Modify  | 0.029273   | 0.029273   | 0.029273   |   0.0 |  0.71

Other   |            | 0.002351   |            |       |  0.06

Pair    | 12.594     | 12.594     | 12.594     |   0.0 | 90.27

Bond    | 7.3e-05    | 7.3e-05    | 7.3e-05    |   0.0 |  0.00

Kspace  | 0.56296    | 0.56296    | 0.56296    |   0.0 |  4.04

Neigh   | 0.65858    | 0.65858    | 0.65858    |   0.0 |  4.72

Comm    | 0.019093   | 0.019093   | 0.019093   |   0.0 |  0.14

Output  | 0.055025   | 0.055025   | 0.055025   |   0.0 |  0.39

Modify  | 0.057276   | 0.057276   | 0.057276   |   0.0 |  0.41

Other   |            | 0.004003   |            |       |  0.03



Nlocal:    4500 ave 4500 max 4500 min

Histogram: 1 0 0 0 0 0 0 0 0 0
@@ -127,10 +168,10 @@ Histogram: 1 0 0 0 0 0 0 0 0 0 
Neighs:    2.60198e+06 ave 2.60198e+06 max 2.60198e+06 min

Histogram: 1 0 0 0 0 0 0 0 0 0



Total # of neighbors = 2601984

Ave neighs/atom = 578.219

Total # of neighbors = 2601983

Ave neighs/atom = 578.218

Ave special neighs/atom = 2

Neighbor list builds = 4

Dangerous builds = 1



Total wall time: 0:00:04
 
Total wall time: 0:00:15

examples/USER/tally/log.21Aug15.force.4→examples/USER/tally/log.12Jun17.force.4

+177 −0
Original line number Diff line number Diff line 
LAMMPS (21 Aug 2015-ICMS)

  using 1 OpenMP thread(s) per MPI task

LAMMPS (19 May 2017)



units		real

atom_style	full
@@ -50,10 +49,39 @@ fix 1 all shake 0.0001 20 0 b 1 a 1 
  1500 = # of frozen angles

fix		2 all nvt temp 300.0 300.0 100.0



# make certain that shake constraints are satisfied

run 0 post no

PPPM initialization ...

WARNING: Using 12-bit tables for long-range coulomb (../kspace.cpp:321)

  G vector (1/distance) = 0.218482

  grid = 15 15 15

  stencil order = 5

  estimated absolute RMS force accuracy = 0.0319435

  estimated relative force accuracy = 9.61968e-05

  using double precision FFTs

  3d grid and FFT values/proc = 3380 960

Neighbor list info ...

  update every 1 steps, delay 10 steps, check yes

  max neighbors/atom: 2000, page size: 100000

  master list distance cutoff = 14

  ghost atom cutoff = 14

  binsize = 7, bins = 6 6 6

  1 neighbor lists, perpetual/occasional/extra = 1 0 0

  (1) pair lj/cut/coul/long, perpetual

      attributes: half, newton on

      pair build: half/bin/newton

      stencil: half/bin/3d/newton

      bin: standard

Per MPI rank memory allocation (min/avg/max) = 10.6 | 10.61 | 10.61 Mbytes

Step Temp E_pair E_mol TotEng Press 

       0            0   -16692.358            0   -16692.358   -1289.8319 

Loop time of 4.5e-06 on 4 procs for 0 steps with 4500 atoms





group		one molecule 1 2

6 atoms in group one



# the following section shows equivalences between using the pe/tally compute and other computes and thermo keywords

# the following section shows equivalences between using the force/tally compute and other computes and thermo keywords



# compute pairwise force between two molecules and everybody

compute		fpa one group/group all pair yes kspace no boundary no
@@ -69,17 +97,19 @@ variable fpa equal sqrt(c_fpa[1]*c_fpa[1]+c_fpa[2]*c_fpa[2]+c_fpa[3]*c_fpa[3]) 
variable	for equal sqrt(c_one[1]*c_one[1]+c_one[2]*c_one[2]+c_one[3]*c_one[3])

# round to 10**-10 absolute precision.

variable	ref equal round(1e10*sqrt(c_red[1]*c_red[1]+c_red[2]*c_red[2]+c_red[3]*c_red[3]))*1e-10

variable	all equal round(1e10*c_c2)*1e-10



velocity	all create 300 432567 dist uniform



timestep	2.0



# v_fpa and v_for and c_c1, c_fpa[] and c_one[] should all each have the same value. v_ref and c_c2 should be zero

thermo_style    custom step v_fpa v_for c_c1 c_fpa[1] c_one[1] c_fpa[2] c_one[2] c_fpa[3] c_one[3] v_ref  c_c2

thermo_style    custom step v_fpa v_for c_c1 c_fpa[1] c_one[1] c_fpa[2] c_one[2] c_fpa[3] c_one[3] v_ref v_all

thermo		10



run 50

PPPM initialization ...

WARNING: Using 12-bit tables for long-range coulomb (../kspace.cpp:321)

  G vector (1/distance) = 0.218482

  grid = 15 15 15

  stencil order = 5
@@ -87,38 +117,49 @@ PPPM initialization ... 
  estimated relative force accuracy = 9.61968e-05

  using double precision FFTs

  3d grid and FFT values/proc = 3380 960

WARNING: Compute force/tally only called from pair style (../compute_force_tally.cpp:75)

WARNING: Compute force/tally only called from pair style (../compute_force_tally.cpp:75)

WARNING: Compute force/tally only called from pair style (../compute_force_tally.cpp:77)

WARNING: Compute force/tally only called from pair style (../compute_force_tally.cpp:77)

Neighbor list info ...

  2 neighbor list requests

  update every 1 steps, delay 10 steps, check yes

  max neighbors/atom: 2000, page size: 100000

  master list distance cutoff = 14

  ghost atom cutoff = 14

  binsize = 7 -> bins = 6 6 6

Memory usage per processor = 8.16441 Mbytes

Step fpa for c1 fpa[1] one[1] fpa[2] one[2] fpa[3] one[3] ref c2 

       0    22.732789    22.732789    22.732789   -17.068392   -17.068392   -8.8345214   -8.8345214   -12.140878   -12.140878            0            0 

      10    11.736915    11.736915    11.736915   -3.3898298   -3.3898298     9.119272     9.119272   -6.5652948   -6.5652948            0            0 

      20    5.6119761    5.6119761    5.6119761  -0.60028931  -0.60028931   -4.4479886   -4.4479886     3.368876     3.368876            0            0 

      30    17.292617    17.292617    17.292617    6.1793856    6.1793856   -10.593927   -10.593927    12.190919    12.190919            0            0 

      40    18.664226    18.664226    18.664226    5.4725079    5.4725079    -6.933046    -6.933046    16.441955    16.441955            0            0 

      50    12.130282    12.130282    12.130282   -1.0321244   -1.0321244    8.0032646    8.0032646   -9.0568326   -9.0568326            0            0 



Loop time of 1.13658 on 4 procs for 50 steps with 4500 atoms

100.0% CPU use with 4 MPI tasks x 1 OpenMP threads

Performance: 7.602 ns/day  3.157 hours/ns  43.991 timesteps/s



MPI task timings breakdown:

  binsize = 7, bins = 6 6 6

  2 neighbor lists, perpetual/occasional/extra = 1 1 0

  (1) pair lj/cut/coul/long, perpetual

      attributes: half, newton on

      pair build: half/bin/newton

      stencil: half/bin/3d/newton

      bin: standard

  (2) compute group/group, occasional, copy from (1)

      attributes: half, newton on

      pair build: copy

      stencil: none

      bin: none

Per MPI rank memory allocation (min/avg/max) = 11.58 | 11.59 | 11.6 Mbytes

Step v_fpa v_for c_c1 c_fpa[1] c_one[1] c_fpa[2] c_one[2] c_fpa[3] c_one[3] v_ref v_all 

       0      22.7331      22.7331      22.7331   -17.068295   -17.068295   -8.8348335   -8.8348334   -12.141369   -12.141369            0            0 

      10    11.736901    11.736901    11.736901   -3.3897029   -3.3897029    9.1193856    9.1193856   -6.5651786   -6.5651786            0            0 

      20    5.6120339    5.6120339    5.6120339  -0.60046861  -0.60046861   -4.4481306   -4.4481306    3.3687528    3.3687528            0            0 

      30     17.29261     17.29261     17.29261     6.179302     6.179302   -10.593979   -10.593979    12.190906    12.190906            0            0 

      40    18.664433    18.664433    18.664433    5.4727782    5.4727782   -6.9329319   -6.9329319    16.442148    16.442148            0            0 

      50    12.130407    12.130407    12.130407   -1.0321196   -1.0321196    8.0035558    8.0035558   -9.0567428   -9.0567428            0            0 

Loop time of 4.31614 on 4 procs for 50 steps with 4500 atoms



Performance: 2.002 ns/day, 11.989 hours/ns, 11.584 timesteps/s

31.6% CPU use with 4 MPI tasks x no OpenMP threads



MPI task timing breakdown:

Section |  min time  |  avg time  |  max time  |%varavg| %total

---------------------------------------------------------------

Pair    | 0.85795    | 0.89088    | 0.93636    |   3.0 | 78.38

Bond    | 3.4571e-05 | 4.4644e-05 | 5.4598e-05 |   0.1 |  0.00

Kspace  | 0.059847   | 0.1051     | 0.1384     |   8.9 |  9.25

Neigh   | 0.085891   | 0.085954   | 0.086      |   0.0 |  7.56

Comm    | 0.01758    | 0.018091   | 0.019178   |   0.5 |  1.59

Output  | 0.013697   | 0.013725   | 0.013805   |   0.0 |  1.21

Modify  | 0.021068   | 0.021137   | 0.021205   |   0.0 |  1.86

Other   |            | 0.001656   |            |       |  0.15

Pair    | 3.5075     | 3.6114     | 3.7489     |   4.7 | 83.67

Bond    | 8.6e-05    | 0.00010525 | 0.000141   |   0.0 |  0.00

Kspace  | 0.2581     | 0.39489    | 0.49723    |  14.2 |  9.15

Neigh   | 0.19826    | 0.19888    | 0.19918    |   0.1 |  4.61

Comm    | 0.034639   | 0.037137   | 0.038938   |   0.9 |  0.86

Output  | 0.025465   | 0.025997   | 0.027558   |   0.6 |  0.60

Modify  | 0.044022   | 0.044175   | 0.044407   |   0.1 |  1.02

Other   |            | 0.003593   |            |       |  0.08



Nlocal:    1125 ave 1148 max 1097 min

Histogram: 1 0 0 1 0 0 0 0 1 1
@@ -127,10 +168,10 @@ Histogram: 1 0 0 1 0 1 0 0 0 1 
Neighs:    650496 ave 675112 max 631353 min

Histogram: 1 0 0 1 1 0 0 0 0 1



Total # of neighbors = 2601984

Ave neighs/atom = 578.219

Total # of neighbors = 2601983

Ave neighs/atom = 578.218

Ave special neighs/atom = 2

Neighbor list builds = 4

Dangerous builds = 1



Total wall time: 0:00:01
 
Total wall time: 0:00:04

examples/USER/tally/log.21Aug15.pe.1→examples/USER/tally/log.12Jun17.pe.1

+71 −31
Original line number Diff line number Diff line 
LAMMPS (21 Aug 2015-ICMS)

  using 1 OpenMP thread(s) per MPI task

LAMMPS (19 May 2017)



units		real

atom_style	full
@@ -50,6 +49,35 @@ fix 1 all shake 0.0001 20 0 b 1 a 1 
  1500 = # of frozen angles

fix		2 all nvt temp 300.0 300.0 100.0



# make certain that shake constraints are satisfied

run 0 post no

PPPM initialization ...

WARNING: Using 12-bit tables for long-range coulomb (../kspace.cpp:321)

  G vector (1/distance) = 0.218482

  grid = 15 15 15

  stencil order = 5

  estimated absolute RMS force accuracy = 0.0319435

  estimated relative force accuracy = 9.61968e-05

  using double precision FFTs

  3d grid and FFT values/proc = 8000 3375

Neighbor list info ...

  update every 1 steps, delay 10 steps, check yes

  max neighbors/atom: 2000, page size: 100000

  master list distance cutoff = 14

  ghost atom cutoff = 14

  binsize = 7, bins = 6 6 6

  1 neighbor lists, perpetual/occasional/extra = 1 0 0

  (1) pair lj/cut/coul/long, perpetual

      attributes: half, newton on

      pair build: half/bin/newton

      stencil: half/bin/3d/newton

      bin: standard

Per MPI rank memory allocation (min/avg/max) = 26.54 | 26.54 | 26.54 Mbytes

Step Temp E_pair E_mol TotEng Press 

       0            0   -16692.358            0   -16692.358   -1289.8319 

Loop time of 1e-06 on 1 procs for 0 steps with 4500 atoms





group		oxy type 1

1500 atoms in group oxy

group		hyd type 2
@@ -88,6 +116,7 @@ thermo 10 


run 50

PPPM initialization ...

WARNING: Using 12-bit tables for long-range coulomb (../kspace.cpp:321)

  G vector (1/distance) = 0.218482

  grid = 15 15 15

  stencil order = 5
@@ -95,38 +124,49 @@ PPPM initialization ... 
  estimated relative force accuracy = 9.61968e-05

  using double precision FFTs

  3d grid and FFT values/proc = 8000 3375

WARNING: Compute pe/tally only called from pair style (../compute_pe_tally.cpp:75)

WARNING: Compute pe/tally only called from pair style (../compute_pe_tally.cpp:75)

WARNING: Compute pe/tally only called from pair style (../compute_pe_tally.cpp:77)

WARNING: Compute pe/tally only called from pair style (../compute_pe_tally.cpp:77)

Neighbor list info ...

  2 neighbor list requests

  update every 1 steps, delay 10 steps, check yes

  max neighbors/atom: 2000, page size: 100000

  master list distance cutoff = 14

  ghost atom cutoff = 14

  binsize = 7 -> bins = 6 6 6

Memory usage per processor = 17.381 Mbytes

Step epa epa E_vdwl vdwl E_coul coul eref pe c2 pair 

       0   -516632.19   -516632.19    3169.9382    3169.9382    46213.889    46213.889    49383.827    49383.827    49383.827    49383.827 

      10   -517027.36   -517027.36    3099.1322    3099.1322     45891.84     45891.84    48990.972    48990.972    48990.972    48990.972 

      20   -516828.06   -516828.06    3101.4321    3101.4321     45884.14     45884.14    48985.572    48985.572    48985.572    48985.572 

      30    -517032.1    -517032.1    3198.5939    3198.5939    45793.571    45793.571    48992.165    48992.165    48992.165    48992.165 

      40   -517095.56   -517095.56    3244.0797    3244.0797    45715.265    45715.265    48959.345    48959.345    48959.345    48959.345 

      50   -517273.54   -517273.54    3274.9142    3274.9142    45665.997    45665.997    48940.911    48940.911    48940.911    48940.911 



Loop time of 4.31105 on 1 procs for 50 steps with 4500 atoms

100.1% CPU use with 1 MPI tasks x 1 OpenMP threads

Performance: 2.004 ns/day  11.975 hours/ns  11.598 timesteps/s



MPI task timings breakdown:

  binsize = 7, bins = 6 6 6

  2 neighbor lists, perpetual/occasional/extra = 1 1 0

  (1) pair lj/cut/coul/long, perpetual

      attributes: half, newton on

      pair build: half/bin/newton

      stencil: half/bin/3d/newton

      bin: standard

  (2) compute group/group, occasional, copy from (1)

      attributes: half, newton on

      pair build: copy

      stencil: none

      bin: none

Per MPI rank memory allocation (min/avg/max) = 29.08 | 29.08 | 29.08 Mbytes

Step c_epa v_epa E_vdwl v_vdwl E_coul v_coul v_eref v_pe c_c2 v_pair 

       0   -516634.27   -516634.27    3169.9427    3169.9427    46212.482    46212.482    49382.425    49382.425    49382.425    49382.425 

      10   -517027.35   -517027.35    3099.1374    3099.1374    45891.866    45891.866    48991.003    48991.003    48991.003    48991.003 

      20   -516828.05   -516828.05    3101.4373    3101.4373    45884.156    45884.156    48985.594    48985.594    48985.594    48985.594 

      30   -517032.07   -517032.07    3198.5951    3198.5951    45793.595    45793.595    48992.191    48992.191    48992.191    48992.191 

      40   -517095.54   -517095.54    3244.0771    3244.0771    45715.292    45715.292    48959.369    48959.369    48959.369    48959.369 

      50    -517273.5    -517273.5    3274.9097    3274.9097    45666.025    45666.025    48940.935    48940.935    48940.935    48940.935 

Loop time of 15.3339 on 1 procs for 50 steps with 4500 atoms



Performance: 0.563 ns/day, 42.594 hours/ns, 3.261 timesteps/s

32.0% CPU use with 1 MPI tasks x no OpenMP threads



MPI task timing breakdown:

Section |  min time  |  avg time  |  max time  |%varavg| %total

---------------------------------------------------------------

Pair    | 3.5071     | 3.5071     | 3.5071     |   0.0 | 81.35

Bond    | 0.00025034 | 0.00025034 | 0.00025034 |   0.0 |  0.01

Kspace  | 0.19991    | 0.19991    | 0.19991    |   0.0 |  4.64

Neigh   | 0.31459    | 0.31459    | 0.31459    |   0.0 |  7.30

Comm    | 0.010338   | 0.010338   | 0.010338   |   0.0 |  0.24

Output  | 0.24722    | 0.24722    | 0.24722    |   0.0 |  5.73

Modify  | 0.029466   | 0.029466   | 0.029466   |   0.0 |  0.68

Other   |            | 0.002182   |            |       |  0.05

Pair    | 13.432     | 13.432     | 13.432     |   0.0 | 87.60

Bond    | 0.000365   | 0.000365   | 0.000365   |   0.0 |  0.00

Kspace  | 0.581      | 0.581      | 0.581      |   0.0 |  3.79

Neigh   | 0.66081    | 0.66081    | 0.66081    |   0.0 |  4.31

Comm    | 0.019908   | 0.019908   | 0.019908   |   0.0 |  0.13

Output  | 0.57731    | 0.57731    | 0.57731    |   0.0 |  3.76

Modify  | 0.058515   | 0.058515   | 0.058515   |   0.0 |  0.38

Other   |            | 0.003889   |            |       |  0.03



Nlocal:    4500 ave 4500 max 4500 min

Histogram: 1 0 0 0 0 0 0 0 0 0
@@ -135,10 +175,10 @@ Histogram: 1 0 0 0 0 0 0 0 0 0 
Neighs:    2.60198e+06 ave 2.60198e+06 max 2.60198e+06 min

Histogram: 1 0 0 0 0 0 0 0 0 0



Total # of neighbors = 2601984

Ave neighs/atom = 578.219

Total # of neighbors = 2601983

Ave neighs/atom = 578.218

Ave special neighs/atom = 2

Neighbor list builds = 4

Dangerous builds = 1



Total wall time: 0:00:04
 
Total wall time: 0:00:16

examples/USER/tally/log.21Aug15.pe.4→examples/USER/tally/log.12Jun17.pe.4

+71 −31

File changed and moved.

Preview size limit exceeded, changes collapsed.

examples/USER/tally/log.21Aug15.stress.1→examples/USER/tally/log.12Jun17.stress.1

+56 −33

File changed and moved.

Preview size limit exceeded, changes collapsed.

CRA Git | Maintained and supported by SUSTech CRA and CCSE