NERSC logo National Energy Research Scientific Computing Center
  A DOE Office of Science User Facility
  at Lawrence Berkeley National Laboratory
 
PackagePlatformVersionModule Docs
nag bassi 32-bit 21 nag_32 NERSCVendor
nag bassi 64-bit Mark 20 nag_64/20 NERSCVendor
nag bassi 64-bit Mark 21 nag_64 NERSCVendor
nag bassi SMP 64-bit 2.0 nag_smp NERSCVendor
nag jacquard 21 nag NERSCVendor
(*) Denotes limited support

NAG Parallel Library (MPI-based version)

The Numerical Algorithms Group (NAG) Parallel math and stat library provides routines for dedicated parallel machines and clusters of workstations or PCs. Please note NAG parallel library and NAG Fortran library are two different products.

The NAG Parallel Library makes use of the Basic Linear Algebra Communication Subprograms (BLACS) which uses MPI as its message-passing kernels and includes the following areas: Sparse Matrix Solvers, Dense and Banded Linear Equations, Eigenvalue and singular value problems, Quadrature, Discrete Fourier Transforms in one, two and three dimensions, Random Number Generation, Matrix Generation and Distribution and Optimization.

How to use

%module load nag_par   
%mpxlf90 $NAG_PAR ... 

Example Programs

NAG Parallel example programs can be accessed by the command nagmpiexample. For example,

%nagmpiexample f04ebfp

will copy the source code of the specified example program and its data into the files f04ebfpe.f and f04ebfpe.d in the current directory and then compile, link, and execute the example and write its results to stdout. Further usage details may be obtained by using the command nagmpiexample without arguments.

Documentation

There are NERSC-specific notes and HTML/pdf documentation at the NAG site. For an overview of other information related to NAG, one can visit the NAG Home Page

Subroutines naming convention

The Library is divided into chapters, each devoted to a branch of numerical analysis or statistics. Each chapter has a three-character name and a title, e.g.

  D01 - Quadrature

Note: The chapters and their names are based on the ACM modified SHARE classification index.

All routines in the Library have either six-character names ending in "F" or seven-character names ending in "FP" and beginning with the characters of the chapter name, e.g.

	D01DAFP

Note that the second and third characters are digits, not letters; e.g. 0 is the digit zero, not the letter O.

The routines with six-character names are identical to routines with the same name, in the NAG Fortran library.


LBNL Home
Page last modified: Fri, 28 Apr 2006 23:58:36 GMT
Page URL: http://www.nersc.gov/nusers/resources/software/libs/math/nag_par/
Web contact: webmaster@nersc.gov
Computing questions: consult@nersc.gov

Privacy and Security Notice
DOE Office of Science