| Titre : | Efficient Approach for the Virtual machine placement in Cloud data center |
| Auteurs : | Abderrahim Baghdad, Auteur ; Samir Setaouti, Directeur de thèse |
| Type de document : | texte manuscrit |
| Editeur : | Université mustapha stambouli de Mascara:Faculté des sciences exactes, 2023 |
| ISBN/ISSN/EAN : | SE02348T |
| Format : | 52p. / couv.ill. / 29cm |
| Accompagnement : | disque optique numérique (CD-ROM) |
| Langues: | Anglais |
| Résumé : |
The optimization of VMs placement in Cloud data centers is a topic that has been the subject of several researches and works due to its importance. Cloud providers seek to derive maximum benefit by optimizing the resources use and at the same time minimizing the energy consumed. By trying to reduce the number of PMs used to minimize the energy cost, performance degradation issues can arise and therefore have SLA violations that penalize the provider. In our end-of-studies work, we proposed an improved genetic algorithm to optimize the placement of VMs while trying to find a compromise between the minimization of the energy consumed and the minimization of the risk of SLA violations. The proposed algorithm uses three different techniques for the generation of the initial population namely: Random, Random first fit and Random best fit. The proposed algorithm relies on directed crossover and mutation to ensure the feasibility of individuals with respect to CPU usage. We have also introduced a migration-based repair function to attempt to meet CPU usage thresholds. The individuals generated are evaluated by a fitness function which takes into consideration the energy consumed and the risk of SLA violation. To test the performance of the proposed algorithm, we implemented a simulator for the placement of VMs. The experiments were carried out according to several parameters concerning the CPU utilization thresholds and the configuration of the proposed genetic algorithm. The results obtained showed the effectiveness of the proposed algorithm against the MBFD technique in terms of energy minimization as well as in the reduction of the risk of SLA violation. In future work, we plan to take other criteria into consideration such as network traffic saturation, cooling energy. We also plan to switch to grouped genetic algorithms to take advantage of the parallelism offered by Cloud platforms. |
Exemplaires (1)
| Code-barres | Cote | Support | Localisation | Section | Disponibilité |
|---|---|---|---|---|---|
| SE02348T | INF934 | Livre audio | Bibliothèque des Sciences Exactes | 7-Mémoires Master | Consultation sur place Exclu du prêt |
Aucun avis, veuillez vous identifier pour ajouter le vôtre !
Accueil
