Compiling and testing LAPACK on Mills

The NAG sites has a collection of examples to test LAPACK drivers. A driver routine will call the necessary lower level routines to solve one particular problem. For example, a real linear least square problem is solved by the dgels driver. Driver routines may not be all LAPACK libraries, but you can download drivers from netlib driver rouines. The source of the driver routine is useful for learning how to use the lower level routines.

Each example in the NAG collection has a source file, a input file, and a output result file, which should match your result.

dels-ex.f - The Fortran 77 source file
ddels-ex.d - The input data file to be read from unit 5 (standard input)
dgels-ex.r - Should match the output on unit 6 (standard output)

You can use wget to get these with the script:

if [ ! -f "dgels-ex.f" ]; then
  wget http://www.nag.com/lapack-ex/examples/source/dgels-ex.f 
  wget http://www.nag.com/lapack-ex/examples/data/dgels-ex.d 
  wget http://www.nag.com/lapack-ex/examples/results/dgels-ex.r 
else 
  touch "dgels-ex.f"
fi

You can just type the three wget commands in your terminal window, but it is a good idea to save them in a file for later reference. In this case, you should enclose them in a conditional if statement to avoid downloading a file you already have.

The Intel Composer XE Suites comes installed with a Fortran compiler the MKL library. Use VALET, vpkg_versions intel. to find the latest version installed on Mills - Version 2013 (2.144).

Versions: The newest package name is Intel Composer XE SP1 and both update 1 and 2 are installed on Mills. You can get the Package name and infer the update number from VALET, but you may also need the version of the compiler and the version of the LAPACK interfaces supported in the MKL component of the package. See the Intel Development site. From the release notes for Linux Fortran 2013sp1,

 Update 2 - February 2014
   Intel Fortran Compiler updated to 14.0.2
   Intel Math Kernal Library updated to 11.1 Update 2

and the details on the main product page for MKL 11.1,

 LAPACK 3.4.1 interfaces and enhancements

Assuming you have the dgels-ex.f Fortran 77 source file, use the VALET and complile commands to compile the source file to an executable that links with the MKL library.

vpkg_devrequire intel/14.0.2-64bit
ifort -mkl dgels-ex.f -o dgels-ex

The ifort compiler has an -mkl optimization flag, and from the man page or ifort –help

   -mkl[=<arg>]
          link to the Intel(R) Math Kernel Library (Intel(R) MKL) and bring
          in the associated headers
            parallel   - link using the threaded Intel(R) MKL libraries. This
                         is the default when -mkl is specified
            sequential - link using the non-threaded Intel(R) MKL libraries

            cluster    - link using the Intel(R) MKL Cluster libraries plus
                         the sequential Intel(R) MKL libraries

Using make: This is a simple compiler command, but you may want to get prepared for more complicated projects, with multiple source files, libraries and compiler flags. Here are the commands to run the same compile command (using make's implicit rules)

vpkg_devrequire intel/14.0.2-64bit
export FC=ifort
export FFLAGS=-mkl
make dgels-ex

The ifort compiler with flag -mkl will compile and link to the threaded MKL libraries. Thus you should test in the threaded parallel environment, and export the number of slots to the MKL_NUM_THREAD environment variable.

test.qs

#$ -N dgels-ex
#$ -pe threads 4
 
echo "--- Set environment ---"
source /opt/shared/valet/docs/valet.sh
vpkg_require intel/14.0.2-64bit
 
echo ""
echo "--- Run Test with $NSLOTS threads ---"
export MKL_NUM_THREADS=$NSLOTS
time ./$JOB_NAME < $JOB_NAME.d
 
echo ""
echo "--- Compare Results ---"
cat $JOB_NAME.r

--- Set environment ---
WARNING: 'gcc' was not found
Adding package `intel/2013-2.144-64bit` to your environment

--- Run Test with 4 threads ---
 DGELS Example Program Results
 
 Least squares solution
      1.5339     1.8707    -1.5241     0.0392
 
 Square root of the residual sum of squares
      2.22E-02

real	0m0.966s
user	0m0.003s
sys	0m0.031s

--- Compare Results ---
 DGELS Example Program Results

 Least squares solution
      1.5339     1.8707    -1.5241     0.0392

 Square root of the residual sum of squares
      2.22E-02

WARNING: 'gcc' was not found: The release notes for this version of intel composer suite indicate that GNU gdk is included for debugging, which requires gcc. This warning can be ignored if you are not debugging with gdk in the batch script. You will not get this warning on the head node, since the system version of gcc will always be found in your path.

If your are debugging on the compute nodes or want to remove the warning, add

vpkg_require gcc/4.6

before the intel vpkg_require command in you batch script file.

Sub-second timing results are not reliable. This test is not a benchmark and was meant to show that you can compile an link a program to read the data file, call the LAPACK routine dgels, and reproduce the correct results.

This example used the default parallel MKL libraries. The LAPACK library is a collection of routines, which parallelize nicely (for large problems), and MKL is an optimized multi-threaded library. For large probrems you get the best performance with the default. However, there are three important considerations when using MKL.

Programs with small arrays will not benefit from the multi-threaded library, and may suffer a bit from the system overhead of maintaining multiple threads.
Sequential programs are better suited for running simultaneous instances. You could run 12 copies of the program on the same node with better throughput when you compile them to be sequential. (Too many threads on the same node will contend for limited resources)
You may be able to take control of the parallelism in your program with OPENMP compiler directions. This is easiest if you using the single threaded MKL in your parallel regions. See recommended settings for calling intel MKL routines from multi threaded applications.

The AMD core math library (ACML) is from AMD developers, and is thus a good chioce form the Mills chip set. Use VALET, vpkg_versions acml. to find the latest version installed on Mills - 5.3.0.

Versions: From the release notes in the file /opt/shared/ACML/5.3.0/ReleaseNotes

 New features of release 5.3.0 of ACML
   Updated the LAPACK code to version 3.4.0.

Compiling and testing LAPACK on Mills

Getting the example files

Compiling with intel and mkl library

VALET and ifort

qsub file to test

Test result output

Sequential vs parallel

Compiling with PGI and ACML library

hpc documentation