Click on a topic to skip to that section.
The IMSL library contains over one thousand functions. These routines provide easy access to high quality implementations of numerical methods in mathematics and statistics.
Mathematical routines include: linear algebra, eigensystem analysis, interpolation and approximation, integration and differentiation, solving differential equations, Fourier and Laplace transforms, finding solutions of nonlinear equations, optimization, and special functions.
Statistical routines include: regression correlation, analysis of variance, categorical and discrete data analysis, nonparametric statistics, tests of goodness of fit and randomness, time series analysis and forecasting, covariance structures and factor analysis, discriminant analysis, cluster analysis, sampling, survival analysis, life testing, and reliability, multidimensional scaling, density and hazard estimation, probability distribution functions and inverses, and random number generation.
The IMSL functions are called like ordinary functions, then the IMSL libraries are linked into the executable when the calling program is loaded.
IMSL is proprietary software and cannot be transferred to a machine that is not licensed to run it.
The IMSL C Numerical libraries are now included as part of Visual Numerics' Computational Technology Toolkit (CTT). IMSL C Numerical libraries V5.5 comes as part of CTT 6.0 and is available on the ITS Unix servers running on the Solaris operating system: ccwf.cc.utexas.edu and uts.cc.utexas.edu. If you want to use IMSL on your departmental machines then you must obtain a license for it. The University of Texas does not have a source license for the IMSL library. If you are interested in obtaining either a source license or a binary license contact:
Visual Numerics Coporate Headquarters 1300 W. Sam Houston Pkwy S. Suite 150 Houston, Texas 77042 USA 800-364-8880 713-784-3131
They can also be reached by e-mail at, info@houston.vin.com.
source /usr/local/imsl/CTT6.0/ctt/bin/cttsetup.cshThe command above can be put at the end of your .cshrc (or .profile) so that it will be executed automatically for you. It is important that you issue this command only after you have set your environment variables such as your PATH and MANPATH. Once you have executed the script, use:
cc [compiler options] $CFLAGS filename $LINK_CNL
to compile and load the program in filename using the default
(shared) version of the IMSL library. It is also possible to explicitly
specify the use of the static or shared library. To load the static
or shared versions of the IMSL libraries, respectively, use:
cc [compiler options] $CFLAGS filename $LINK_CNL_STATIC
or
cc [compiler options] $CFLAGS filename $LINK_CNL_SHARED
To generate an executable eigentest that uses the shared
version of the IMSL libraries, compile the example file eigentest.c
with:
cc -o eigentest $CFLAGS eigentest.c $LINK_CNL
You may be using a completely different machine or operating system, but this example might still prove to be helpful. The following program uses the IMSL routines imsl_f_eig_sym and imsl_f_write_matrix. The discussion below assumes that the contents of the file eigentest.c are:
#include <imsl.h>
main()
{
int n = 3;
float a[] = {7.0, -8.0, -8.0,
-8.0, -16.0, -18.0,
-8.0, -18.0, 13.0};
float *eval;
/* Compute eigenvalues */
eval = imsl_f_eig_sym(n, a, 0);
/* Print eigenvalues */
imsl_f_write_matrix ("Eigenvalues", 1, 3, eval, 0);
}
These are pointers to information on the Internet that you may find helpful. Be aware, though, that because these resources are maintained at remote sites, the information may not always apply to you, and may not always be available.
Getting help from the Applications Consultants
SSC provides consulting services for a wide range of applications by electronic mail, telephone, and office visits. For an overview of the consulting services available to you, see SSC Consulting Services.
Assistance for IMSL is provided by SSC Mathematical Support, who you can reach by electronic mail at:
math@ssc.utexas.edu
Please include a description of the problem including any particular IMSL routines you might be trying to use, a log of the relevant parts of your compilation or execution, and any data your program might be using. Also, if possible include a campus or other local telephone number where a consultant can reach you.
Additional information about this or other mathematical and statistical software can be obtained from the SSC Consulting Home Page.