Use LAMMPS to perform a manufacturing simulation - Elastic High Performance Computing

This topic describes how to run LAMMPS in an Elastic High Performance Computing (E-HPC) cluster to perform a manufacturing simulation based on the 3d Lennard-Jones melt model and visualize the simulation result.

Background information

Large-scale Atomic/Molecular Massively Parallel Simulator (LAMMPS) is a classical molecular dynamics program. It has potentials for solid-state materials (metals, semiconductors), soft matter (biomolecules, polymers), and coarse-grained or mesoscopic systems.

High-performance and flexible E-HPC clusters provide auxiliary analysis for complex engineering and mechanics. You can take advantage of these benefits to run large amounts of simulations and optimize product structures and performance. More and more software packages are used for manufacturing simulations.

Preparations

Create an E-HPC cluster. For more information, see Create a cluster by using the wizard.

The following parameter configurations are used in this example.

Parameter	Description
Hardware settings	Set the Deploy Mode to Standard. Specify two management nodes, one compute node, and one logon node. Select an instance type that has 32 or more vCPUs as the compute node. For example, you can select ecs.c7.8xlarge. Note This example is designed for an application with relatively low computing requirements. Adjust the number of compute nodes based on the computing requirements of your application.
Software settings	Deploy a CentOS 7.6 public image and the PBS scheduler. Turn on VNC.

Create a cluster user. For more information, see Create a user.
The user is used to log on to the cluster, compile software, and submit jobs. The following settings are used in this example:
- Username: testuser
- User group: sudo permission group
Install software. For more information, see Install software.
Install the following software:
- lammps-openmpi V31Mar17
- openmpi V1.10.7
- VMD V1.9.3

Step 1: Connect to the cluster

Connect to the cluster by using one of the following methods. This example uses testuser as the username. After you connect to the cluster, you are automatically logged on to the /home/testuser.

Use an E-HPC client to log on to a cluster
The scheduler of the cluster must be PBS. Make sure that you have downloaded and installed an E-HPC client and deployed the environment required for the client. For more information, see Deploy an environment for an E-HPC client.
1. Start and log on to your E-HPC client.
2. In the left-side navigation pane, click Session Management.
3. In the upper-right corner of the Session Management page, click terminal to open the Terminal window.
Use the E-HPC console to log on to a cluster
1. Log on to the E-HPC console.
2. In the upper-left corner of the top navigation bar, select a region.
3. In the left-side navigation pane, click Cluster.
4. On the Cluster page, find the cluster and click Connect.
5. In the Connect panel, enter a username and a password, and click Connect via SSH.

Step 2: Submit a job

Run the following command to create a sample file named lj.in:

vim lj.in

The sample file:

# 3d Lennard-Jones melt

variable        x index 1
variable        y index 1
variable        z index 1

variable        xx equal 20*$x
variable        yy equal 20*$y
variable        zz equal 20*$z

units           lj
atom_style      atomic

lattice         fcc 0.8442
region          box block 0 ${xx} 0 ${yy} 0 ${zz}
create_box      1 box
create_atoms    1 box
mass            1 1.0

velocity        all create 1.44 87287 loop geom

pair_style      lj/cut 2.5
pair_coeff      1 1 1.0 1.0 2.5

neighbor        0.3 bin
neigh_modify    delay 0 every 20 check no

fix             1 all nve
dump 1 all xyz 100 sample.xyz
run             10000

Run the following command to create a job script file named lammps.pbs:
```
vim lammps.pbs
```
The sample script:
Note
In this example, the compute node has 32 vCPUs and 32 Message Passing Interface (MPI) processes to perform high-performance computing. Configure the number of vCPUs based on the actual specification of compute nodes that you use. The number of vCPUs must be greater than 32.
```
#!/bin/sh
#PBS -l select=1:ncpus=32:mpiprocs=32     
#PBS -j oe

export MODULEPATH=/opt/ehpcmodulefiles/   # The environment variables on which the module command depends.
module load lammps-openmpi/31Mar17
module load openmpi/1.10.7

echo "run at the beginning"
mpirun lmp -in ./lj.in # Use the actual path of the lj.in file. 
```
Run the following command to submit the job:
```
qsub lammps.pbs
```
The following command output is returned, which indicates that the generated job ID is 0.scheduler.
```
0.scheduler
```

Step 3: View the job result

Run the following command to view the result of the job:

cat lammps.pbs.o0

Note

If you do not specify a standard output path for the job, the output file is generated based on scheduler behaviors. By default, the output file is stored in the /home/<username>/ directory. For example, the output of this example is stored in the /home/testuser/lammps.pbs.o0 directory.

The following code provides an example of the expected returned output.

......
Per MPI rank memory allocation (min/avg/max) = 3.777 | 3.801 | 3.818 Mbytes
Step Temp E_pair E_mol TotEng Press 
       0         1.44   -6.7733681            0   -4.6134356   -5.0197073 
   10000   0.69814375   -5.6683212            0   -4.6211383   0.75227555 
Loop time of 9.81493 on 32 procs for 10000 steps with 32000 atoms

Performance: 440145.641 tau/day, 1018.856 timesteps/s
97.0% CPU use with 32 MPI tasks x no OpenMP threads

MPI task timing breakdown:
Section |  min time  |  avg time  |  max time  |%varavg| %total
---------------------------------------------------------------
Pair    | 6.0055     | 6.1975     | 6.3645     |   4.0 | 63.14
Neigh   | 0.90095    | 0.91322    | 0.92938    |   0.9 |  9.30
Comm    | 2.1457     | 2.3105     | 2.4945     |   6.9 | 23.54
Output  | 0.16934    | 0.1998     | 0.23357    |   4.3 |  2.04
Modify  | 0.1259     | 0.13028    | 0.13602    |   0.8 |  1.33
Other   |            | 0.06364    |            |       |  0.65

Nlocal:    1000 ave 1022 max 986 min
Histogram: 5 3 6 3 4 4 2 2 1 2
Nghost:    2705.62 ave 2733 max 2668 min
Histogram: 1 1 0 3 7 5 4 5 4 2
Neighs:    37505 ave 38906 max 36560 min
Histogram: 7 3 2 4 5 2 3 3 2 1

Total # of neighbors = 1200161
Ave neighs/atom = 37.505
Neighbor list builds = 500
Dangerous builds not checked
Total wall time: 0:00:10

Use VNC to view the result of the job.
1. Enable VNC.
  Note
  Make sure that the ports required by VNC are enabled for the security group to which the cluster belongs. When you use the console, the system automatically enables the port 12016. When you use the client, you need to enable the ports manually. Port 12017 allows only one user to open the VNC Viewer window. If multiple users need to open the VNC Viewer window, you need to enable the corresponding number of ports, starting from port 12017.
  - Use the client
    1. In the left-side navigation pane, click Session Management.
    2. In the upper-right corner of the Session Management page, click VNC to open VNC Viewer.
  - Use the console
    1. In the left-side navigation pane of the E-HPC console, click Cluster.
    2. On the Cluster page, select a cluster. Choose More > VNC.
    3. Use VNC to remotely connect to a visualization service. For more information, see Use VNC to manage a visualization service.
2. In the Virtualization Service dialog box of the cloud desktop, choose Application > System Tools > Terminal.
3. Run the /opt/vmd/1.9.3/vmd to open Visual Molecular Dynamics (VMD).
4. On the VMD Main page, choose File > New Molecule....
5. Click the Browse... button and select the sample.xyz file.
  Note
  The sample.xyz file is stored in the /home/testuser/sample.xyz directory.
6. Click Load. You can view the result of the job in the VMD 1.9.3 OpenGL Display window.