#########################################
# Grid NAS Parallel Benchmarks          #
# F77                                   #
# Revision 3.1-SHF-SER                  #
# NASA Ames Research Center             #
# npb@nas.nasa.gov                      #
# http://www.nas.nasa.gov/Software/NPB/ #
#########################################


This README file explains how to build the Grid NAS Parallel Benchmark 
suite.  Please read the technical report NAS-02-005 well. 

NOTE: this reference implementation does not exploit concurrency.
The NPB codes used are the serial versions, and the NGB graphs are
executed serially as well. A single shared file system is assumed.

1. Edit the site- and machine-specific data in config/make.def
   Several sample versions are included in subdirectory NAS.samples
   of the config directory so you may be able to copy one. A clean 
   version is in config/make.def.template.

2. Each benchmark comes in 4 sizes (classes), ranked in order of
   increasing data set sizes and computational work involved: S, W, A, B.
   Since Fortran 77 doesn't have dynamic memory allocation, the class 
   must be specified at compile time.

   To compile a given benchmark for a specific class, type:

       make benchmark-name CLASS={S,W,A,B}

   For instance, to create a class A version of the VP benchmark, type:

         make VP CLASS=A

   Note: the process-intensive "Rapid Fire" (parameter "-RF") variation 
   of the grid benchmarks only use Class S executables. To build these
   benchmarks, always select "CLASS=S", regardless of which class is
   specified during execution.

   All executables related to the benchmark that has been built are
   placed in the subdirectory "bin" of the distribution (or in the directory 
   BINDIR specified in make.def, if you have defined it). The whole
   benchmark is executed using a Korn shell script, located in the root 
   directory of the distribution (name is all UPPERCASE). It
   takes the class of the whole benchmark as a parameter. Other parameters
   are -v for verbose execution, -u for unformatted data files, and -RF
   for the Rapid Fire variation of the benchmarks.
   By default, the script runs with minimal information written to standard 
   output, and using ASCII files for exchanging data between tasks. 
   Example run:

         VP A -u -v

3. It is possible to make all benchmarks of a particular class at the
   same time by typing: 

         make all CLASS=<class>

3a. When building a 

4. It is also possible to build single executables of a particular class,
   for example:

         make BT CLASS=A

5. If the benchmark completes it will automatically remove any temporary 
   files that were created. However, if it crashes, no complete cleanup will 
   be done, and the user should remove any newly created files, if desired. 
   Leftover temporary files will not interfere with subsequent benchmark 
   runs, but may clog the file system.