Accelerating an algorithm for perishable inventory control on heterogeneous platforms

A. Gutierrez-Alcoba, G. Ortega, E.M.T. Hendrix, I. García

Research output: Contribution to journalArticleAcademicpeer-review

3 Citations (Scopus)

Abstract

This paper analyses and evaluates parallel implementations of an optimization algorithm for perishable inventory control problems. This iterative algorithm has high computational requirements when solving large problems. Therefore, the use of parallel and distributed computing reduces the execution time and improves the quality of the solutions. This work investigates two implementations on heterogeneous platforms: (1) a MPI-PTHREADS version; and (2) a multi-GPU version. A comparison of these implementations has been carried out. Experimental results show the benefits of using parallel and distributed codes to solve this kind of problems. Furthermore, the distribution of the workload among the available processing elements is a challenging problem. This distribution of tasks can be modelled as a Bin-Packing problem. This implies that the selection of the set of tasks assigned to every processing element requires the design of a heuristic capable of efficiently balancing the workload statically with no significant overhead. This heuristic has been used for the parallel implementations of the optimization for perishable inventory control problem.
LanguageEnglish
Pages12-18
JournalJournal of Parallel and Distributed Computing
Volume104
DOIs
Publication statusPublished - Jun 2017

Fingerprint

Inventory control
Inventory Control
Parallel Implementation
Workload
Control Problem
Distributed computer systems
Bins
Parallel processing systems
Processing
Heuristics
Parallel and Distributed Computing
Bin Packing Problem
Balancing
Execution Time
Iterative Algorithm
Optimization Algorithm
Imply
Optimization
Evaluate
Requirements

Cite this

@article{af3e3653a52f477583af0282e606a0f9,
title = "Accelerating an algorithm for perishable inventory control on heterogeneous platforms",
abstract = "This paper analyses and evaluates parallel implementations of an optimization algorithm for perishable inventory control problems. This iterative algorithm has high computational requirements when solving large problems. Therefore, the use of parallel and distributed computing reduces the execution time and improves the quality of the solutions. This work investigates two implementations on heterogeneous platforms: (1) a MPI-PTHREADS version; and (2) a multi-GPU version. A comparison of these implementations has been carried out. Experimental results show the benefits of using parallel and distributed codes to solve this kind of problems. Furthermore, the distribution of the workload among the available processing elements is a challenging problem. This distribution of tasks can be modelled as a Bin-Packing problem. This implies that the selection of the set of tasks assigned to every processing element requires the design of a heuristic capable of efficiently balancing the workload statically with no significant overhead. This heuristic has been used for the parallel implementations of the optimization for perishable inventory control problem.",
author = "A. Gutierrez-Alcoba and G. Ortega and E.M.T. Hendrix and I. Garc{\'i}a",
year = "2017",
month = "6",
doi = "10.1016/j.jpdc.2016.12.021",
language = "English",
volume = "104",
pages = "12--18",
journal = "Journal of Parallel and Distributed Computing",
issn = "0743-7315",
publisher = "Elsevier",

}

Accelerating an algorithm for perishable inventory control on heterogeneous platforms. / Gutierrez-Alcoba, A.; Ortega, G.; Hendrix, E.M.T.; García, I.

In: Journal of Parallel and Distributed Computing, Vol. 104, 06.2017, p. 12-18.

Research output: Contribution to journalArticleAcademicpeer-review

TY - JOUR

T1 - Accelerating an algorithm for perishable inventory control on heterogeneous platforms

AU - Gutierrez-Alcoba, A.

AU - Ortega, G.

AU - Hendrix, E.M.T.

AU - García, I.

PY - 2017/6

Y1 - 2017/6

N2 - This paper analyses and evaluates parallel implementations of an optimization algorithm for perishable inventory control problems. This iterative algorithm has high computational requirements when solving large problems. Therefore, the use of parallel and distributed computing reduces the execution time and improves the quality of the solutions. This work investigates two implementations on heterogeneous platforms: (1) a MPI-PTHREADS version; and (2) a multi-GPU version. A comparison of these implementations has been carried out. Experimental results show the benefits of using parallel and distributed codes to solve this kind of problems. Furthermore, the distribution of the workload among the available processing elements is a challenging problem. This distribution of tasks can be modelled as a Bin-Packing problem. This implies that the selection of the set of tasks assigned to every processing element requires the design of a heuristic capable of efficiently balancing the workload statically with no significant overhead. This heuristic has been used for the parallel implementations of the optimization for perishable inventory control problem.

AB - This paper analyses and evaluates parallel implementations of an optimization algorithm for perishable inventory control problems. This iterative algorithm has high computational requirements when solving large problems. Therefore, the use of parallel and distributed computing reduces the execution time and improves the quality of the solutions. This work investigates two implementations on heterogeneous platforms: (1) a MPI-PTHREADS version; and (2) a multi-GPU version. A comparison of these implementations has been carried out. Experimental results show the benefits of using parallel and distributed codes to solve this kind of problems. Furthermore, the distribution of the workload among the available processing elements is a challenging problem. This distribution of tasks can be modelled as a Bin-Packing problem. This implies that the selection of the set of tasks assigned to every processing element requires the design of a heuristic capable of efficiently balancing the workload statically with no significant overhead. This heuristic has been used for the parallel implementations of the optimization for perishable inventory control problem.

U2 - 10.1016/j.jpdc.2016.12.021

DO - 10.1016/j.jpdc.2016.12.021

M3 - Article

VL - 104

SP - 12

EP - 18

JO - Journal of Parallel and Distributed Computing

T2 - Journal of Parallel and Distributed Computing

JF - Journal of Parallel and Distributed Computing

SN - 0743-7315

ER -