JASPA is a benchmark containing Fortran90, C and Java codes for testing
the I/O and compute performance of sparse matrix multiplication
kernels,which appear frequently in large scale scientific and engineering
applications. It can be used to compare the performance
of Java with Fortran90 and C on your chosen platform.
You can find benchmarking results on the Pentium II,
the Sun Ultra 80, the IBM PowerPC and Power3 in the report:
"Comparing the performance of
JAVA with Fortran and C for numerical computing"
Performance Comparison of Java with Fortran90/C
The following is a quick comparison of Java with Fortran90 and C using the
"memplus"
sparse matrix which
comes with the benchmark suite. We measure the floating point
performance (in Mflops/s). Note however
that floating point operations form only part
of the overall computing effort in the benchmark, so the
Mflops/s rate is only meaningful in a relative sense
when comparing the three languages across the platforms.
For more details see here.
Download
JASPA is freely available. Click
to download.
1) Installation
JASPA comes as a gzipped tar file, JASPA.tar.gz, you need to unpack it
by
%gunzip JASPA.tar.gz
%tar xvf JASPA.tar.gz
This will create a JASPA directory, which contains
a number of subdirectories.
2) Compiling
Edit common.make in the SPARSE_BENCH directory to give the correct
Fortran90, C and JAVA compiler flags, you can find some sample in
the MAKE.sample directory, e.g., LINUX.make. Once you have selected the
correct flags, you do
% make
to compile the Fortran90, C and Java codes.
3) Test Matrices
You need to have test sparse matrices in the
MatrixMarket coordinate
format, but with entry values
removed. There are some sample matries in the directory
SPARSE_BENCH/matrices, they are in the MatrixMarket coordinate format,
with the entry values removed. If you want to test on other matrices
from MatrixMarket, you need to down load the matrices into the
SPARSE_BENCH/matrices directory, and remove the entry values by
% MatrixMarket2bench "matrix file" > "new matrix file"
4) Running the Benchmark
To run Fortran90, C and JAVA benchmark on a matrix "memplus.mtx" in the
SPARSE_BENCH/matrices directory, you need to cd to that directory
and do
% bench_f90 memplus.mtx
% bench_c memplus.mtx
% java Spmatmul2 memplus.mtx
The outcomes of running, say, the Java benchmark on a Pentium II under
Linux is:
=========== JAVA benchmark for sparse multiplication ==========
case name = memplus.mtx
problem size = 17758
nz = 126150
time_IO = 1.424 averaged over 2 runs
nz1 = 5121784
time_COMP = 2.969 averaged over 1 runs
Mflops/s = 8.137172111822162
flops/run = 24159264
average entry value = 1.858481341657516
Here the timing is measures over a number of runs so that the total
time over all runs exceed a preset minimum (set to 2.0 second in the
current version). The Mflops/s measures the floating point
performance, although it is only meaningful for relative comparison
across the platforms, since a lot of the computing is spent not in
floating point type operations.
5) Running the Benchmark Automatically
In the SPARSE_BENCH/matrices directory, there is a perl script
"run.perl" in the SPARSE_BENCH/util directory which will automatically
run the Fortran90, C and Java codes on all the matrices of the form "*.mtx"
in that directory. You need to make sure that the path for perl is
correct at line one of this script. Then in this directory, do
% run.perl
The output will be in "out.f90", "out.c" and "out.java".
|
