On parallelizing dual decomposition in stochastic integer programming

For stochastic mixed-integer programs, we revisit the dual decomposition algorithm of Carøe and Schultz from a computational perspective with the aim of its parallelization. We address an important bottleneck of parallel execution by identifying a formulation that permits the parallel solution of th...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:Operations research letters Ročník 41; číslo 3; s. 252 - 258
Hlavní autoři: Lubin, Miles, Martin, Kipp, Petra, Cosmin G., Sandıkçı, Burhaneddin
Médium: Journal Article
Jazyk:angličtina
Vydáno: Elsevier B.V 01.05.2013
Elsevier
Témata:
Bundle methods
Column generation
Computation
Decomposition
Dual decomposition
Integer programming
Mixed-integer programming
Operations research
Optimization
Parallel computing
Parallel processing
Solvers
Stochastic programming
Stochasticity
Dual decomposition
Column generation
Parallel computing
Bundle methods
Mixed-integer programming
Stochastic programming
ISSN:0167-6377, 1872-7468
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:For stochastic mixed-integer programs, we revisit the dual decomposition algorithm of Carøe and Schultz from a computational perspective with the aim of its parallelization. We address an important bottleneck of parallel execution by identifying a formulation that permits the parallel solution of the master program by using structure-exploiting interior-point solvers. Our results demonstrate the potential for parallel speedup and the importance of regularization (stabilization) in the dual optimization. Load imbalance is identified as a remaining barrier to parallel scalability.
Bibliografie:ObjectType-Article-2
SourceType-Scholarly Journals-1
ObjectType-Feature-1
content type line 23
ISSN:0167-6377
1872-7468
DOI:10.1016/j.orl.2013.02.003