Using Genetic Algorithms to Map Hard Real-Time NoC-based Systems

Adrian Racu; Leandro Soares Indrusiak

Using Genetic Algorithms to Map Hard Real-Time NoC-based Systems

Computer Science

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abstract

This paper investigates the effectiveness of genetic algorithms (GAs) for static task scheduling in wormhole Network-on-Chip-based systems. The overall objective was to get the application model mapped onto the architecture so that all tasks and communication meet their deadlines. Inter-task communication is accounted for by using analytical methods. The GA explores both the mapping of tasks as well as the priority ordering of the task set. A novel fitness function was developed and found to perform better than existing functions.

Original language	English
Title of host publication	2012 7th International Workshop on Reconfigurable Communication-centric Systems-on-Chip (ReCoSoC)
Editors	LS Indrusiak, G Gogniat, N Voros
Place of Publication	NEW YORK
Publisher	IEEE
Pages	-
Number of pages	8
ISBN (Print)	978-1-4673-2572-1
Publication status	Published - 2012
Event	7th International Workshop on Reconfigurable Communication-centric Systems-on-Chip (ReCoSoC) - York Duration: 9 Jul 2012 → 11 Jul 2012

Conference

Conference	7th International Workshop on Reconfigurable Communication-centric Systems-on-Chip (ReCoSoC)
City	York
Period	9/07/12 → 11/07/12

Cite this

@inproceedings{560353135b71472f9444ec92788fff78,

title = "Using Genetic Algorithms to Map Hard Real-Time NoC-based Systems",

abstract = "This paper investigates the effectiveness of genetic algorithms (GAs) for static task scheduling in wormhole Network-on-Chip-based systems. The overall objective was to get the application model mapped onto the architecture so that all tasks and communication meet their deadlines. Inter-task communication is accounted for by using analytical methods. The GA explores both the mapping of tasks as well as the priority ordering of the task set. A novel fitness function was developed and found to perform better than existing functions.",

author = "Adrian Racu and Indrusiak, {Leandro Soares}",

year = "2012",

language = "English",

isbn = "978-1-4673-2572-1",

pages = "--",

editor = "LS Indrusiak and G Gogniat and N Voros",

booktitle = "2012 7th International Workshop on Reconfigurable Communication-centric Systems-on-Chip (ReCoSoC)",

publisher = "IEEE",

address = "United States",

note = "7th International Workshop on Reconfigurable Communication-centric Systems-on-Chip (ReCoSoC) ; Conference date: 09-07-2012 Through 11-07-2012",

}

Racu, A & Indrusiak, LS 2012, Using Genetic Algorithms to Map Hard Real-Time NoC-based Systems. in LS Indrusiak, G Gogniat & N Voros (eds), 2012 7th International Workshop on Reconfigurable Communication-centric Systems-on-Chip (ReCoSoC). IEEE, NEW YORK, pp. -, 7th International Workshop on Reconfigurable Communication-centric Systems-on-Chip (ReCoSoC), York, 9/07/12.

TY - GEN

T1 - Using Genetic Algorithms to Map Hard Real-Time NoC-based Systems

AU - Racu, Adrian

AU - Indrusiak, Leandro Soares

PY - 2012

Y1 - 2012

N2 - This paper investigates the effectiveness of genetic algorithms (GAs) for static task scheduling in wormhole Network-on-Chip-based systems. The overall objective was to get the application model mapped onto the architecture so that all tasks and communication meet their deadlines. Inter-task communication is accounted for by using analytical methods. The GA explores both the mapping of tasks as well as the priority ordering of the task set. A novel fitness function was developed and found to perform better than existing functions.

AB - This paper investigates the effectiveness of genetic algorithms (GAs) for static task scheduling in wormhole Network-on-Chip-based systems. The overall objective was to get the application model mapped onto the architecture so that all tasks and communication meet their deadlines. Inter-task communication is accounted for by using analytical methods. The GA explores both the mapping of tasks as well as the priority ordering of the task set. A novel fitness function was developed and found to perform better than existing functions.

M3 - Conference contribution

SN - 978-1-4673-2572-1

SP - -

BT - 2012 7th International Workshop on Reconfigurable Communication-centric Systems-on-Chip (ReCoSoC)

A2 - Indrusiak, LS

A2 - Gogniat, G

A2 - Voros, N

PB - IEEE

CY - NEW YORK

T2 - 7th International Workshop on Reconfigurable Communication-centric Systems-on-Chip (ReCoSoC)

Y2 - 9 July 2012 through 11 July 2012

ER -