Realization of Parallel Ant Colony Algorithm Based on TBB Multi-core Platform

Author

Li, Ni ; Gao, Dongdong ; Gong, Guanghong ; Chen, Zheng

Author_Institution

Sch. of Autom. Sci. & Electr. Eng., Beijing Univ. of Astronaut. & Aeronaut., Beijing, China

Volume

1

fYear

2010

fDate

16-18 July 2010

Firstpage

177

Lastpage

180

Abstract

TBB (Thread Building Blocking) is currently a representative parallel computing platform of multi-core processors. The ant colony algorithm is used to solve combinatorial optimization problem of discrete-time systems. With the expansion of the problem scale, it often results in rapid increase of calculation. Based on TBB a parallel ant colony algorithm was researched and developed to improve the efficiency of the algorithm. TBB parallel multi-core technology and implementation techniques of the parallel ant colony algorithm based on TBB are introduced in detail. This algorithm was applied to resolve large-scale TSP (traveling salesman problem). Experiment results showed that the parallel optimization progress is flexible to operate and multi-core resources can be made full use of to greatly improve algorithm efficiency. And it provided possibility for real-time calculation of large-scale combinatorial optimization engineering problems.

Keywords

discrete time systems; optimisation; parallel algorithms; travelling salesman problems; TBB multicore platform; combinatorial optimization problem; discrete time system; implementation technique; large scale TSP; multicore processor; parallel ant colony algorithm; parallel computing; parallel optimization; thread building blocking; traveling salesman problem; Algorithm design and analysis; Cities and towns; Instruction sets; Optimization; Parallel algorithms; Resource management; ant colony algorithm; multi-core parallel computing; thread building blocking; traveling salesman problem;

fLanguage

English

Publisher

ieee

Conference_Titel

Information Technology and Applications (IFITA), 2010 International Forum on

Conference_Location

Kunming

Print_ISBN

978-1-4244-7621-3

Electronic_ISBN

978-1-4244-7622-0

Type

conf

DOI

10.1109/IFITA.2010.143

Filename

5635135