Semidefinite Programming Packages
This page is continuously updated!
Last modified: March 12, 2000

Name of the Package Author(s) Purpose
CSDP 2.3 Brian Borchers
New Mexico Tech Mathematics Faculty
Department of Mathematics
Socorro, NM 87801
Office: Weir Hall 158
Telephone: 505-835-5813
Fax: 505-835-5366
borchers@nmt.edu
CSDP is a library of routines that implements a predictor corrector variant of the semidefinite programming algorithm of Helmberg, Rendl, Vanderbei, and Wolkowicz.
The main advantages of this code are that it is written to be used as a callable subroutine, it is written in C for efficiency, it makes effective use of sparsity in the constraint matrices, and that it includes support for linear inequality constraints in addition to linear equality constraints.
CUTSDP Stefan E. Karisch
Carmen Systems AB, Gothenburg, Sweden.
E-mail: Stefan.Karisch@carmen.se
CUTSDP is a package of C programs containing an implementation of a cutting plane approach based on semidefinite programming. Currently, there are three applications implemented: max-cut, graph bisection, and graph equipartition.
Manual:
Stefan E. Karisch. CUTSDP - A Toolbox for a Cutting-Plane Approach Based on Semidefinite Programming.
Technical Report IMM-REP-1998, Department of Mathematical Modelling,
Technical University of Denmark, June 17, 1998.
MAXDET Shao-Po Wu
Information Systems Laboratory,
Stanford University, Stanford, CA 94305
E-mail: clive@stanford.edu
Lieven Vandenberghe
UCLA Electrical Engineering Department
68-119 Engineering IV
Los Angeles, CA 90095-1594
E-mail: vandenbe@ee.ucla.edu
Stephen P. Boyd
Stanford University, Packard 264, Stanford, CA 94305
E-mail: boyd@stanford.edu
Software for determinant maximization problems. Implementation of a long-step path-following method for determinant maximization problems.
Includes full C-source (with calls to LAPACK), which can be used directly or via matlab mex file interfaces, matlab examples, and documentation.

Manual of MAXDET: User's Guide, Alpha Version. May 24, 1996.
SDPA Katsuki Fujisawa
Masakazu Kojima
Department of Mathematical and Computing Sciences,
Tokyo Institute of Technology,
2-12-1 Oh-Okayama, Meguro-ku,
Tokyo 152-0033, Japan
User's Manual, ps.Z-file (ftp).
See also:
K. Fujisawa, M. Fukuda, M. Kojima and K. Nakata.
"Numerical Evaluation of SDPA",
Research Report B-330, Department of Mathematical and Computing Sciences, Tokyo Institute of
Technology, Oh-Okayama, Meguro-ku, Tokyo 152, September 1997.
ps.Z-file (ftp) or
dvi.Z-file (ftp)
SDPHA Nathan Brixius
517 South Linn St. #1
Iowa City, IA 52240
E-mail: brixius@cs.uiowa.edu
Rongqin Sheng
MCS Division, Argonne National Laboratory
Florian A. Potra
e-mail: potra@math.umbc.edu
SDPHA: A Matlab package for semidefinite programming.
MATLAB code for predictor-corrector algorithms for semidefinite programming using the homogeneous formulation
README file for SDPHA v3.0.
User guide SDPHA (updated 6/29/99).
Technical paper. (updated 6/29/99)
SDPPACK Farid Alizadeh,
RUTCOR, Rutgers University, New Brunswick, NJ.
E-mail: alizadeh@rutcor.rutgers.edu
Jean-Pierre A. Haeberly,
Department of Mathematics, Fordham University, Bronx, NY.
E-mail: haeberly@murray.fordham.edu.
Madhu V. Nayakkankuppam,
Courant Institute of Mathematical Sciences, New York University, NY.
E-mail: madhu@cs.nyu.edu.
Michael L. Overton,
Courant Institute of Mathematical Sciences, New York University, NY.
E-mail: overton@cs.nyu.edu
Stefan Schmieta
RUTCOR, Rutgers University, New Brunswick, NJ.
E-mail: schmieta@rutcor.rutgers.edu
SDPpack Version 0.9 beta runs under Matlab 5.0. This version extends the previous release for semidefinite programming (SDP) to mixed semidefinite-quadratic-linear programs (SQLP), i.e. linear optimization problems over a product of semidefinite cones, quadratic cones and the nonnegative orthant. Together, these cones make up all possible homogeneous self-dual cones over the reals. The main routine implements a primal-dual Mehrotra predictor-corrector scheme based on the XZ+ZX search direction.
User Guide
User Guide PostScript.
User Guide DVI.
SDPSOL Shao-Po Wu
Information Systems Laboratory,
Stanford University, Stanford, CA 94305
E-mail: clive@stanford.edu
Stephen P. Boyd
Stanford University, Packard 264, Stanford, CA 94305
E-mail: boyd@stanford.edu
SDPSOL is a parser/solver for SDP and MAXDET problems with matrix structure.
MAXDET problems have the form:
minimize c^Tx - log det G(x)
subject to G(x) > 0, F(x) > 0,
where G(x)>0 and F(x)>0 are linear matrix inequality (LMI) constraints.
Two important special cases are SDP (when G(x)=1) and analytic centering (when c=0 and F(x)=1).
MAXDET (and SDP) problems arise in control, statistics, computational geometry, and information and communication theory.
In many cases the optimization variables have matrix structure, which makes it tedius in practice to put the problem in the form above. SDPSOL automates this task by allowing the user to specify (and solve) MAXDET (or SDP) problems in a format close to its natural mathematical description. SDPSOL parses problems expressed in the SDPSOL language, solves them using an interior-point method, and reports the results in a convenient form.
User's Guide PostScript.
User's Cuide PDF.
SDPSOL: a parser/solver for SDP and MAXDET problems with matrix structure.
In: Recent Advances in LMI Methods for Control,
Edited by L. El Ghaoui and S.-I. Niculescu, SIAM, 1999.
SDPT3 Toh Kim Chuan,
Department of Mathematics
National University of Singapore
10 Kent Ridge Crescent
Singapore 119260
Singapore
Email: mattohkc@math.nus.edu.sg
Michael J. Todd
School of Operations Research and Industrial Engineering
229 Frank H.T. Rhodes Hall
Cornell University, Ithaca, NY 14853
Email: miketodd@orie.cornell.edu or miketodd@cs.cornell.edu
Reha Tütüncü
Department of Mathematical Sciences
6113 Wean Hall
Carnegie Mellon University
Pittsburgh, PA 15213.
Matlab package for Infeasible path-following and homogeneous self-dual algorithms for solving standard SDP (possibly with complex data).
Sparsity in the data is exploited whenever possible.

K.C. Toh, M.J. Todd, and R.H. Tutuncu,
SDPT3 --- a Matlab software package for semidefinite programming, version 2.1. (User's Guide PostScript)
SeDuMi Jos F. Sturm,
Faculty of Economics and Business Administration,
Department of Quantitative Economics,
Maastricht University, P.O. Box 616,
6200 MD Maastricht,
The Netherlands.
E-mail: j.sturm@ke.unimaas.nl
Matlab toolbox for solving optimization problems over symmetric cones, i.e. it allows not only for linear constraints, but also quasiconvex-quadratic constraints and positive semi-definiteness constraints. Complex valued entries are allowed. Both symbolic and numerical reordering schemes, balancing speed/accuracy performance. Sophisticated dense column handling, using Goldfarb-Scheinberg product form idea.
User's Guide: Using SeDuMi 1.02 a MATLAB toolbox for optimization over symmetric cones.
SP Lieven Vandenberghe
UCLA Electrical Engineering Department
68-119 Engineering IV
Los Angeles, CA 90095-1594
E-mail: vandenbe@ee.ucla.edu
Stephen P. Boyd
Stanford University, Packard 264, Stanford, CA 94305
E-mail: boyd@stanford.edu
Brien Alkire
UCLA Electrical Engineering Department
School of Engineering and Applied Science
University of California, Los Angeles
Los Angeles, CA 90095.
Email: brien@alkires.com
Implementation of Nesterov and Todd's primal-dual potential reduction method for semidefinite programming.
The code is written in C/C++ with calls to BLAS and LAPACK.
Includes full C-source (with calls to LAPACK), which can be used directly or via matlab mex file interfaces, matlab examples, and documentation.

Lieven Vandenberghe and Stephen Boyd,
Semidefinite Programming SIAM Review 38 (1996), 49-95.