Kanad Hpc Application Reference Manual
Total Page:16
File Type:pdf, Size:1020Kb
KANAD HPC APPLICATION REFERENCE MANUAL Page 1 of 28 TABLE OF CONTENTS 1. INTRODUCTION ............................................................................................................................................ 4 2. COMPLIERS & LIBRARIES ............................................................................................................................... 4 2.1. INTEL MKL ............................................................................................................................................... 4 2.2. FFTW ....................................................................................................................................................... 4 2.3. OPENMPI ................................................................................................................................................ 5 2.4. INTEL® MPI LIBRARY ................................................................................................................................ 6 2.5. GCC ......................................................................................................................................................... 6 2.6. INTEL 13.0.1 ............................................................................................................................................. 7 2.7. ATLAS (AUTOMATICALLY TUNED LINEAR ALGEBRA SOFTWARE) ................................................................. 7 2.8. BLAS ........................................................................................................................................................ 7 2.9. GSL ......................................................................................................................................................... 7 2.10. LIBXC ....................................................................................................................................................... 8 2.11. NETCDF ................................................................................................................................................... 8 2.12. HDF5 ....................................................................................................................................................... 8 2.13. CFITSIO .................................................................................................................................................... 8 2.14. PFFT ........................................................................................................................................................ 8 2.15. ETSF ........................................................................................................................................................ 9 2.16. SPARSKIT ................................................................................................................................................. 9 3. APPLICATIONS............................................................................................................................................ 10 3.1. GROMACS ............................................................................................................................................. 10 3.2. DL-POLY ................................................................................................................................................ 12 3.3. ESPRESSO .............................................................................................................................................. 13 3.4. GAUSSIAN ............................................................................................................................................. 15 Page 2 of 28 3.5. LAMMPS ............................................................................................................................................... 17 3.6. MPI-BLAST ............................................................................................................................................. 18 3.7. NAMD ................................................................................................................................................... 19 3.8. NWCHEM .............................................................................................................................................. 20 3.9. OCTOPUS .............................................................................................................................................. 21 3.10. PLUMED-AMBER .................................................................................................................................... 23 3.11. PLUMED-GROMACS ............................................................................................................................... 24 3.12. TINKER .................................................................................................................................................. 25 3.13. CP2K ..................................................................................................................................................... 25 3.14. CPMD .................................................................................................................................................... 26 3.15. CAMB .................................................................................................................................................... 26 3.16. TMOLEX ................................................................................................................................................ 27 3.17. COSMOMC ............................................................................................................................................ 27 3.18. DALTON ................................................................................................................................................ 27 3.19. HEALPIX ................................................................................................................................................ 28 3.20. MYDYNAMIX ......................................................................................................................................... 28 Page 3 of 28 1. Introduction 2. Compliers & Libraries 2.1. Intel MKL Intel® Math Kernel Library (Intel® MKL) offers highly optimized extensively threaded math routines for scientific, engineering, and financial applications that require maximum performance. It includes the deployment of BLAS, BLACS, LAPACK and SCALAPCK routines that are highly optimized for Intel processors, and provides significant performance improvements over alternative implementations. Features: Includes both C and FORTRAN interfaces. Extremely optimized for current multi-core x86 platform. SCALAPAC can provide significant performance improvements over the standard NETLIB implementation. Utilize multi-dimensional FFT routines (1D through 7D) with a modern, easy to use C and Fortran interfaces. It also provides compatibility with the FFTW 2.x and 3.x interfaces making it easy for current FFTW users to plug Intel MKL into their existing applications. Supports distributed memory cluster with the same API enabling to improve performance by distributing the work over large number of processors with minimal efforts. Installation Path: /opt/intel/mkl/ 2.2. FFTW FFTW is a C subroutine library for computing the discrete Fourier transform (DFT) in one or more dimensions, of arbitrary input size, and of both real and complex data (as well as of even/odd data, i.e. the discrete cosine/sine transforms or DCT/DST). It can compute transforms of real- and complex-valued arrays of arbitrary size and dimension in O(n log n) time. Different version are installed as per the requirement of various applications. Installation Path: /opt/apps/libs/fftw/3.3.3/intel/ /opt/apps/FFTW Strictly Confidential Page 4 of 28 2.3. OpenMPI Installation Path: /opt/mpi/openmpi/1.6.2/intel/ /opt/mpi/openmpi/1.4.5/intel/ To compile Parallel – C program : /opt/mpi/openmpi/1.6.2/intel/bin/mpicc C++ program : /opt/mpi/openmpi/1.6.2/intel/bin/mpicxx F77 program : /opt/mpi/openmpi/1.6.2/intel/bin/mpif77 F90 program : /opt/mpi/openmpi/1.6.2/intel/bin/mpif90 To run openMPI parallel jobs: 1. Assuming that executable name is a.out. 2. $/opt/mpi/openmpi/1.6.2/intel/bin/mpirun -np 16 ./a.out (This will run with 4 mpi processes) 3. To define hosts; create a hostfile stating ib node names associating with number of slots: (e.g. this file will define 4 nodes with 4, 2, 6, and 4 slots of ibn1, ibn2, ibn3, and ibn4 respectively). $ vi ibhosts ibn1 slots=4 ibn2 slots=2 ibn3 slots=6 ibn4 slots=4 4. Run the job with “-hostfile <hostfile name>” option: $/opt/mpi/openmpi/1.6.2/intel/bin/mpirun -np 16 -hostfile ibhosts ./a.out Page 5 of 28 2.4. Intel® MPI Library Intel MPI 4.1 focuses on making applications perform better on Intel® architecture- based clusters—implementing the high performance Message Passing Interface Version 2.2 specification on multiple fabrics. It enables you to quickly deliver maximum end user performance even if you change or upgrade to new interconnects, without requiring changes to the software or operating environment. Installation Path: /opt/intel/impi/4.1.0.024/intel64 To compile Parallel – C program : /opt/intel/impi/4.1.0.024/intel64/bin/mpiicc C++ program : /opt/ intel/impi/4.1.0.024/intel64/bin/mpicxx F77 program : /opt/ intel/impi/4.1.0.024/intel64/bin/mpif77 F90 program