Mapping DAG to CGRA Using Min-Max Ant Colony System

Author

Li Zhou ; Hengzhu Liu

Author_Institution

Sch. of Comput. Sci., Nat. Univ. of Defense Technol., Changsha, China

fYear

2013

fDate

10-12 Oct. 2013

Firstpage

520

Lastpage

523

Abstract

Coarse-grained reconfigurable array (CGRA) architecture has become popular because of its performance and flexibility. The efficiency of CGRA relies on an efficient application mapping algorithm to exploit parallelisms. In this paper, we formalized the temporal mapping problem and proposed the min-max ant colony system (MMAS) algorithm for CGRA mapping. Further optimization of MMAS is studied to reduce mapping time while maintaining the quality of solutions. Comparisons with other heuristic algorithms show that our approach obtains better results in less mapping time.

Keywords

ant colony optimisation; directed graphs; minimax techniques; reconfigurable architectures; CGRA efficiency; CGRA mapping; DAG; MMAS algorithm; MMAS optimization; application mapping algorithm; coarse-grained reconfigurable array architecture; directed acyclic graph; heuristic algorithms; mapping time reduction; minmax ant colony system algorithm; temporal mapping problem; Algorithm design and analysis; Cities and towns; Computer architecture; Heuristic algorithms; Kernel; Optimization; Signal processing algorithms; CGRAt; MMAS; application mapping;

fLanguage

English

Publisher

ieee

Conference_Titel

Cyber-Enabled Distributed Computing and Knowledge Discovery (CyberC), 2013 International Conference on

Conference_Location

Beijing

Type

conf

DOI

10.1109/CyberC.2013.96

Filename

6685737