Instruction-set architecture exploration of VLIW aSIPs using a genetic algorithm

R. Jordans; L. Jozwiak; H. Corporaal

doi:10.1109/MECO.2014.6862720

Instruction-set architecture exploration of VLIW aSIPs using a genetic algorithm

R. Jordans, L. Jozwiak, H. Corporaal

Electronic Systems

Onderzoeksoutput: Hoofdstuk in Boek/Rapport/Congresprocedure › Conferentiebijdrage › Academic › peer review

14 Citaten (Scopus)

Samenvatting

Genetic algorithms are commonly used for automatically solving complex design problem because exploration using genetic algorithms can consistently deliver good results when the algorithm is given a long enough run-time. However, the exploration time for problems with huge design spaces can be very long, often making exploration using a genetic algorithm practically infeasible. In this work, we present a genetic algorithm for exploring the instruction-set architecture of VLIW ASIPs and demonstrate its effectiveness by comparing it to two heuristic algorithms. We present several optimizations to the genetic algorithm configuration, and demonstrate how caching of intermediate compilation and simulation results can reduce the exploration time by an order of magnitude.

Originele taal-2	Engels
Titel	Proceedings on the 3rd Mediterranean Conference on Embedded Computing (MECO), 15-19 June 2014, Budva, Montenegro
Plaats van productie	Piscataway
Uitgeverij	Institute of Electrical and Electronics Engineers
Pagina's	32-35
ISBN van geprinte versie	978-1-4799-4827-7
DOI's	https://doi.org/10.1109/MECO.2014.6862720
Status	Gepubliceerd - 2014
Evenement	3rd Mediterranean Conference on Embedded Computing, MECO 2014 - Budva, Montenegro Duur: 15 jun. 2014 → 19 jun. 2014 Congresnummer: 3

Congres

Congres	3rd Mediterranean Conference on Embedded Computing, MECO 2014
Verkorte titel	MECO 2014
Land/Regio	Montenegro
Stad	Budva
Periode	15/06/14 → 19/06/14
Ander	MECO 2014

Toegang tot document

10.1109/MECO.2014.6862720

Citeer dit

@inproceedings{64a35e92443444e8904b115e86c26943,

title = "Instruction-set architecture exploration of VLIW aSIPs using a genetic algorithm",

abstract = "Genetic algorithms are commonly used for automatically solving complex design problem because exploration using genetic algorithms can consistently deliver good results when the algorithm is given a long enough run-time. However, the exploration time for problems with huge design spaces can be very long, often making exploration using a genetic algorithm practically infeasible. In this work, we present a genetic algorithm for exploring the instruction-set architecture of VLIW ASIPs and demonstrate its effectiveness by comparing it to two heuristic algorithms. We present several optimizations to the genetic algorithm configuration, and demonstrate how caching of intermediate compilation and simulation results can reduce the exploration time by an order of magnitude.",

author = "R. Jordans and L. Jozwiak and H. Corporaal",

year = "2014",

doi = "10.1109/MECO.2014.6862720",

language = "English",

isbn = "978-1-4799-4827-7",

pages = "32--35",

booktitle = "Proceedings on the 3rd Mediterranean Conference on Embedded Computing (MECO), 15-19 June 2014, Budva, Montenegro",

publisher = "Institute of Electrical and Electronics Engineers",

address = "United States",

note = "3rd Mediterranean Conference on Embedded Computing, MECO 2014, MECO 2014 ; Conference date: 15-06-2014 Through 19-06-2014",

}

Jordans, R, Jozwiak, L & Corporaal, H 2014, Instruction-set architecture exploration of VLIW aSIPs using a genetic algorithm. in Proceedings on the 3rd Mediterranean Conference on Embedded Computing (MECO), 15-19 June 2014, Budva, Montenegro. Institute of Electrical and Electronics Engineers, Piscataway, blz. 32-35, 3rd Mediterranean Conference on Embedded Computing, MECO 2014, Budva, Montenegro, 15/06/14. https://doi.org/10.1109/MECO.2014.6862720

Instruction-set architecture exploration of VLIW aSIPs using a genetic algorithm. / Jordans, R.; Jozwiak, L.; Corporaal, H.
Proceedings on the 3rd Mediterranean Conference on Embedded Computing (MECO), 15-19 June 2014, Budva, Montenegro. Piscataway: Institute of Electrical and Electronics Engineers, 2014. blz. 32-35.

Onderzoeksoutput: Hoofdstuk in Boek/Rapport/Congresprocedure › Conferentiebijdrage › Academic › peer review

TY - GEN

T1 - Instruction-set architecture exploration of VLIW aSIPs using a genetic algorithm

AU - Jordans, R.

AU - Jozwiak, L.

AU - Corporaal, H.

N1 - Conference code: 3

PY - 2014

Y1 - 2014

N2 - Genetic algorithms are commonly used for automatically solving complex design problem because exploration using genetic algorithms can consistently deliver good results when the algorithm is given a long enough run-time. However, the exploration time for problems with huge design spaces can be very long, often making exploration using a genetic algorithm practically infeasible. In this work, we present a genetic algorithm for exploring the instruction-set architecture of VLIW ASIPs and demonstrate its effectiveness by comparing it to two heuristic algorithms. We present several optimizations to the genetic algorithm configuration, and demonstrate how caching of intermediate compilation and simulation results can reduce the exploration time by an order of magnitude.

AB - Genetic algorithms are commonly used for automatically solving complex design problem because exploration using genetic algorithms can consistently deliver good results when the algorithm is given a long enough run-time. However, the exploration time for problems with huge design spaces can be very long, often making exploration using a genetic algorithm practically infeasible. In this work, we present a genetic algorithm for exploring the instruction-set architecture of VLIW ASIPs and demonstrate its effectiveness by comparing it to two heuristic algorithms. We present several optimizations to the genetic algorithm configuration, and demonstrate how caching of intermediate compilation and simulation results can reduce the exploration time by an order of magnitude.

U2 - 10.1109/MECO.2014.6862720

DO - 10.1109/MECO.2014.6862720

M3 - Conference contribution

SN - 978-1-4799-4827-7

SP - 32

EP - 35

BT - Proceedings on the 3rd Mediterranean Conference on Embedded Computing (MECO), 15-19 June 2014, Budva, Montenegro

PB - Institute of Electrical and Electronics Engineers

CY - Piscataway

T2 - 3rd Mediterranean Conference on Embedded Computing, MECO 2014

Y2 - 15 June 2014 through 19 June 2014

ER -

Instruction-set architecture exploration of VLIW aSIPs using a genetic algorithm

Samenvatting

Congres

Toegang tot document

Vingerafdruk

Citeer dit