SDPA

The NEOS Server offers SDPA for the solution of semidefinite programming problems in sparse SDPA format or in SeDuMi format.

SDPA is a software package for solving semidefinite programs (SDP) with various levels of precision. It is based on a Mehrotra-type predictor-corrector infeasible primal-dual interior-point method. SDPA handles the standard SDP and its dual. It is implemented in C++ utilizing several libraries, in particular the QD package and the GNU multiprecision package.
The variable precision version (SDPA-GMP) is slow and should only be used for small problems. Here are typical times for this solver on a SDPLIB problem (theta3):
double (SDPA): 3 sec, quadruple (SDPA-DD): 211 sec, octuple (SDPA-QD): 2002 sec.

The double precision version runs in parallel on 4 cores.

SDPA was developed by the SDPA group. Source and documentation is available here.

This solver was implemented by Hans Mittelmann and executes at


Using the NEOS Server for SDPA

The user must submit a model in either sparse SDPA or SeDuMi Matlab format to solve a semidefinite programming problem. Examples of models in sparse SDPA format can be found in the SDPLIB library. The same problems in SeDuMi format are here. Other files in this format are at 7th DIMACS Challenge library

If non-standard parameter settings are required, the user may also submit a parameter file (except for the double version). You can download the default parameter file, edit it, and submit it below.



Using the NEOS Server for SDPA/SPARSE_SDPA


Web Submission Form
SDPA data
Enter the complete path to the sparse SDPA format data file
SeDuMi data
Alternatively, enter the complete path to the SeDuMi format data (Matlab binary, containing At b c K).
Precision
Choose the precision (dbl=double (64 bits) [default], quad=quadruple (128 bits), oct=octuple (256 bits), var=variable)




Comments
Additional Settings


E-Mail address:
Please do not click the 'Submit to NEOS' button more than once.