A data replication strategy with tenant performance and provider economic profit guarantees in cloud data centers - IRIT - Institut de Recherche en Informatique de Toulouse Accéder directement au contenu
Article Dans Une Revue Journal of Systems and Software Année : 2020

A data replication strategy with tenant performance and provider economic profit guarantees in cloud data centers

Résumé

Meeting tenant performance requirements through data replication while ensuring an economic profit is very challenging for cloud providers. For this purpose, we propose a data Replication Strategy that satisfies Performance tenant objective and provider profit in Cloud data centers (RSPC). Before the execution of each tenant query Q, data replication is considered only if: (i) the estimated Response Time of Q (RTQ) exceeds a critical RT threshold (per-query replication), or (ii) more often, if RTQ exceeds another (lower) RT threshold for a given number of times (replication per set of queries). Then, a new replica is really created only if a suitable replica placement is heuristically found so that the RT requirement is satisfied again while ensuring an economic profit for the provider. Both the provider's revenues and expenditures are also estimated while penalties and replication costs are taken into account. Furthermore, the replica factor is dynamically adjusted in order to reduce the resource consumption. Compared to four other strategies, RSPC best satisfies the RT requirement under high loads, complex queries and strict RT thresholds. Moreover, penalty and data transfer costs are significantly reduced, which impacts the provider profit.
Fichier principal
Vignette du fichier
Mokadem_26278.pdf (975.08 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)
Loading...

Dates et versions

hal-02878939 , version 1 (23-06-2020)

Identifiants

Citer

Riad Mokadem, Abdelkader Hameurlain. A data replication strategy with tenant performance and provider economic profit guarantees in cloud data centers. Journal of Systems and Software, 2020, 159, pp.110447. ⟨10.1016/j.jss.2019.110447⟩. ⟨hal-02878939⟩
99 Consultations
82 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More