he emergence of large-scale federated Cloud computing environments and of dynamic resource pricing schemes presents interesting saving opportunities for service providers, that could dynamically change the placement of IT service components in order to reduce their bills. However, that calls for smart management solutions able to respond to pricing changes by dynamically reconfiguring IT service component placement in federated Cloud environments so to enforce highlevel business objectives defined by the service providers. This paper proposes a novel adaptive and business-driven IT service component reconfiguration solution based on what-if scenario analysis and on genetic-algorithm optimization. Our solution is able to model complex Cloud computing IT services and to evaluate their performance in a wide range of alternative configurations, by also detecting the optimal placement for their components. The paper presents the experimental evaluation of our framework in a realistic scenario that consists of a 2-tier service architecture with real-world pricing schemes. The results demonstrate the effectiveness of our solution and the suitability of business-driven IT management techniques for the optimal placement of service components in federated Clouds.
Adaptive and business-driven service placement in federated Cloud computing environments
TORTONESI, Mauro
2013
Abstract
he emergence of large-scale federated Cloud computing environments and of dynamic resource pricing schemes presents interesting saving opportunities for service providers, that could dynamically change the placement of IT service components in order to reduce their bills. However, that calls for smart management solutions able to respond to pricing changes by dynamically reconfiguring IT service component placement in federated Cloud environments so to enforce highlevel business objectives defined by the service providers. This paper proposes a novel adaptive and business-driven IT service component reconfiguration solution based on what-if scenario analysis and on genetic-algorithm optimization. Our solution is able to model complex Cloud computing IT services and to evaluate their performance in a wide range of alternative configurations, by also detecting the optimal placement for their components. The paper presents the experimental evaluation of our framework in a realistic scenario that consists of a 2-tier service architecture with real-world pricing schemes. The results demonstrate the effectiveness of our solution and the suitability of business-driven IT management techniques for the optimal placement of service components in federated Clouds.I documenti in SFERA sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
