User Tools

Site Tools


acelab:lammps

Differences

This shows you the differences between two versions of the page.

Link to this comparison view

Next revision
Previous revision
acelab:lammps [2015/09/01 15:37]
mcawood created
acelab:lammps [2021/12/09 16:42] (current)
Line 1: Line 1:
-==General Info==+===General Info===
  
-LAMMPS ("Large-scale Atomic/Molecular Massively Parallel Simulator") is a molecular dynamics program from Sandia National Laboratories.+LAMMPS ("Large-scale Atomic/Molecular Massively Parallel Simulator") is a molecular dynamics program from Sandia National Laboratories.  LAMMPS makes use of MPI for parallel communication and is free, open-source software, distributed under the terms of the GNU General Public License. [[https://en.wikipedia.org/wiki/LAMMPS|source]]
  
-==Useful Links==+===Useful Links===
  
-Developer's Benchmarks [http://lammps.sandia.gov/bench.html] \\ +Developer's Benchmarks [[http://lammps.sandia.gov/bench.html]] \\ 
-HPCAC Best Practices [http://www.hpcadvisorycouncil.com/pdf/LAMMPS_Best_Practice.pdf]+HPCAC Best Practices [[http://www.hpcadvisorycouncil.com/pdf/LAMMPS_Best_Practice.pdf]]
  
-==Installation Guide==+===Installation Guide===
  
 Setup environment: Setup environment:
Line 30: Line 30:
   cd tar   cd tar
  
-===Building the benchmark===+==Building the benchmark==
  
-Download the LAMMPS source (15 May 2015) here: http://www.ace.chpc.ac.za/tars/lammps-stable.tar.gz+Download the LAMMPS source (15 May 2015) here: [[http://www.ace.chpc.ac.za/tars/lammps-stable.tar.gz|lammps-stable.tar.gz]]
  
   tar -xf lammps.stable.tar.gz   tar -xf lammps.stable.tar.gz
Line 38: Line 38:
   cd ../lammps-15May15/   cd ../lammps-15May15/
  
-====Building x86 CPU Benchmark====+=Building x86 CPU Benchmark=
  
 To build the Intel compiled LAMMPS binary, first edit the Makefile: To build the Intel compiled LAMMPS binary, first edit the Makefile:
Line 59: Line 59:
   cp lmp_intel_cpu ../bench   cp lmp_intel_cpu ../bench
  
-====Building CUDA GPU Benchmark====+==Building CUDA GPU Benchmark==
  
 Build this binary using the python script: Build this binary using the python script:
Line 103: Line 103:
   cp lmp_cuda ../   cp lmp_cuda ../
      
-==Benchmark==+===Benchmark===
  
 The stock "3d Lennard-Jones melt" test problem is used as a benchmark for this code. The stock "3d Lennard-Jones melt" test problem is used as a benchmark for this code.
Line 111: Line 111:
 cd ~/LAMMPS/lammps-15May15/bench cd ~/LAMMPS/lammps-15May15/bench
  
-====x86 CPU Benchmark====+A pre-configured input script is available here {{http://wiki.chpc.ac.za/_media/acelab:cpu.tar.gz|cpu.tar.gz}}
  
-A pre-configured input script is available here [[File:Cpu.tar.gz]]+==x86 CPU Benchmark==
  
-For x86 CPU benchmarks, this is '''500K particles per core.'''+For x86 CPU benchmarks, this is **500K particles per core.**
 Therefore, if you with to run the benchmark on 24 x86 cores, a total of 12,000K (500x24) particles is required. Therefore, if you with to run the benchmark on 24 x86 cores, a total of 12,000K (500x24) particles is required.
  
Line 130: Line 130:
 Giving a total particle count of 500K * (6*4*4) = 48x10^6.  Which conforms to the 500K particles per core, (48x10^6 / 96 = 500K). Giving a total particle count of 500K * (6*4*4) = 48x10^6.  Which conforms to the 500K particles per core, (48x10^6 / 96 = 500K).
  
-====CUDA GPU Benchmark====+==CUDA GPU Benchmark==
  
 For the GPU benchmark, the same process is used as above, however with a larger number of particles per GPU. For the GPU benchmark, the same process is used as above, however with a larger number of particles per GPU.
-For GPU benchmarks, the particles per GPU are set as '''8M particles per GPU.'''+For GPU benchmarks, the particles per GPU are set as **8M particles per GPU.**
 Download the pre-configured input file here  Download the pre-configured input file here 
  
/app/dokuwiki/data/attic/acelab/lammps.1441114666.txt.gz · Last modified: 2021/12/09 16:42 (external edit)