Research Reports on Mathematical and Computing Sciences

Abstract. The SDPA (SemiDefinite Programming Algorithm) is a software package for solv-ing semidefinite program (SDP). It is based on a Mehrotra-type predictor-corrector infeasible primal-dual interior-point method. The SDPA handles the standard form SDP and its dual. It is implemented in C++ language utilizing the LAPACK [3] for matrix computation. The SDPA incorporates dynamic memory allocation and deallocation. So, the maximum size of an SDP that can be solved depends on the size of computational memory which user’s computer loads. The SDPA version 6.00 enjoys the following features: • Callable library of the SDPA is available. • Efficient method for computing the search directions when an SDP to be solved is large scale and sparse [5]. • Block diagonal matrix structure and sparse matrix structure in data matrices are available. is incorporated. • Some information on infeasibility of a semidefinite program to be solved is provided. This manual and the SDPA can be found in the WWW sit