Parallel Birge and Qi Factorization of Three-stage Stochastic Linear Programs

• DOI: 10.12694/scpe.v5i3.291
• Publication Date: 2002-01-01
• Authors: Ladislav Halada; Siegfried Benkner; Mária Lucká
• Keywords: Interior point method; Linear algebra; Fortran

Many practical problems, such as portfolio management, problems in manufacturing, transportation or power generation, can be modeled stochastic programs. Stochastic programs provide an effective framework for sequential decision with uncertain data, when uncertainty by a discrete set of scenarios. In this paper we present algorithm solving three-stage linear program based on the Birge and Qi factorization constraint matrix product frame primal-dual path-following interior point method. Moreover, discuss parallelization method distributed-memory machines using Fortran/MPI algebra package LAPACK. Performance experiments 64-processor Beowulf cluster show effectiveness proposed strategy, which exploits two levels parallelism.