Memetic algorithms for parallel code optimization
Discovering the optimum number of processors and the distribution of data on distributed memory parallel computers for a given algorithm is a demanding task. A memetic algorithm (MA) is proposed here to find the best number of processors and the best data distribution method to be used for each stage of a parallel program. Steady state memetic algorithm is compared with transgenerational memetic algorithm using different crossover operators and hill-climbing methods. A self-adaptive MA is also implemented, based on a multimeme strategy. All the experiments are carried out on computationally intensive, communication intensive, and mixed problem instances. The MA performs successfully for the illustrative problem instances. © Springer Science+Business Media, LLC 2007.
| Yazar |
Özcan, E. Onbaşioğlu, E. |
| Yayın Türü | Article |
| Tek Biçim Adres | https://hdl.handle.net/20.500.11831/509 |
| Konu Başlıkları |
Distributed memory parallel computers
Memetic algorithms Parallelizing compilers Search methods |
| Koleksiyonlar |
Araştırma Çıktıları | Ön Baskı | WoS | Scopus | TR-Dizin | PubMed 02- WoS İndeksli Yayınlar Koleksiyonu 03- Scopus İndeksli Yayınlar Koleksiyonu |
| Dergi Adı | International Journal of Parallel Programming |
| Cild | 35 |
| Dergi Sayısı | 1 |
| Sayfalar | 33 - 61 |
| Yayın Tarihi | 2007 |
| Eser Adı [dc.title] | Memetic algorithms for parallel code optimization |
| Yazar [dc.contributor.author] | Özcan, E. |
| Yazar [dc.contributor.author] | Onbaşioğlu, E. |
| Yayıncı [dc.publisher] | Springer New York LLC |
| Yayın Türü [dc.type] | article |
| Özet [dc.description.abstract] | Discovering the optimum number of processors and the distribution of data on distributed memory parallel computers for a given algorithm is a demanding task. A memetic algorithm (MA) is proposed here to find the best number of processors and the best data distribution method to be used for each stage of a parallel program. Steady state memetic algorithm is compared with transgenerational memetic algorithm using different crossover operators and hill-climbing methods. A self-adaptive MA is also implemented, based on a multimeme strategy. All the experiments are carried out on computationally intensive, communication intensive, and mixed problem instances. The MA performs successfully for the illustrative problem instances. © Springer Science+Business Media, LLC 2007. |
| Kayıt Giriş Tarihi [dc.date.accessioned] | 2020-03-17 |
| Yayın Tarihi [dc.date.issued] | 2007 |
| Açık Erişim Tarihi [dc.date.available] | 2020-03-17 |
| Dil [dc.language.iso] | eng |
| Konu Başlıkları [dc.subject] | Distributed memory parallel computers |
| Konu Başlıkları [dc.subject] | Memetic algorithms |
| Konu Başlıkları [dc.subject] | Parallelizing compilers |
| Konu Başlıkları [dc.subject] | Search methods |
| Haklar [dc.rights] | info:eu-repo/semantics/closedAccess |
| ISSN [dc.identifier.issn] | 08857458 |
| Yayının ilk sayfa sayısı [dc.identifier.startpage] | 33 |
| Yayının son sayfa sayısı [dc.identifier.endpage] | 61 |
| Dergi Adı [dc.relation.journal] | International Journal of Parallel Programming |
| Dergi Sayısı [dc.identifier.issue] | 1 |
| Cild [dc.identifier.volume] | 35 |
| Tek Biçim Adres [dc.identifier.uri] | https://hdl.handle.net/20.500.11831/509 |
9
1
30 Eylül 2023 07:08
Tıklayınız
