Time filter

Source Type

Bruneo D.,Messina University | Lhoas A.,College of the Province of Liege | Longo F.,Messina University | Puliafito A.,Messina University
Proceedings - 2013 IEEE 3rd International Conference on Cloud and Green Computing, CGC 2013 and 2013 IEEE 3rd International Conference on Social Computing and Its Applications, SCA 2013 | Year: 2013

Cloud systems represent the new ICT frontier where computing utilities are offered in terms of virtual instances, following the so called as-a-service philosophy. Different commercial solutions have been already put in place but several aspects need to be faced in order to provide high-value services able to meet business requirements. In particular, performance evaluation plays a critical role being strictly related to data center optimization and user satisfaction. Moreover, the use of large data centers able to respond to the high service demand has increased the attention to power efficiency, thus calling for green solutions and energy-aware strategies that jointly consider environmental and economical aspects. In order to design powerful strategies able to meet the quality of service requirements still reducing the energy costs, performance analysis frameworks are needed. In this paper, we present an analytical model, based on stochastic reward nets, that is able to easily implement resource allocation strategies in a green infrastructure as-a-service cloud. The model is organized into layers that represent the virtual resource pool and the physical machines. Different allocation algorithms (i.e., scattering, saturation) have been implemented by properly managing the coordination between the model layers. Numerical results are provided that demonstrate the effectiveness of the proposed approach. © 2013 IEEE.

Masy G.,College of the Province of Liege | Andre P.,University of Liege
HVAC and R Research | Year: 2012

This study is related to a research program intended to develop a very friendly and transparent energy evaluation tool, easy to use by designers, other project partners, and administration (in Belgium). The simulation tool is developed on EES solver. The building envelope model involves a dynamic model composed of resistances and capacities associated to a ventilation model able to assess infiltration/exfiltration air flows through the envelope leakages, as well as natural ventilation air flows combined with fan air flows. An analytical study is presented in this article: it is based on a set of comparative simulations performed on a fictitious office building already used as a test case in other projects of the International Energy Agency. All components of final and primary energy consumption are identified in these simulations in order to obtain a complete breakdown of energy consumptions. Special attention is paid to ventilation (effectiveness, impact on heating/cooling demands, and fan consumption) and to lighting (combination between natural and artificial lighting and corresponding consumption), in relationship with occupants comfort requirement, which appears as having a rather strong impact on energy consumption. Various control strategies are considered in order to improve the energy efficiency without affecting the comfort in terms of indoor air quality and illumination level. © 2012 ASHRAE.

Bruneo D.,Messina University | Lhoas A.,College of the Province of Liege | Longo F.,Messina University | Puliafito A.,Messina University
IEEE Transactions on Parallel and Distributed Systems | Year: 2015

Following the as-a-service philosophy, a cloud service provider offers computing utilities in the form of virtual resources instantiated on top of a physical infrastructure. In order to meet business requirements still providing high-quality services, performance evaluation needs to be carefully carried out with the aim of optimizing data center utilization and increasing user satisfaction. In this context, power efficiency plays a critical role pushing service providers towards the application of innovative green strategies. In this paper, we present an analytical framework, based on stochastic reward nets, that allows to evaluate different resource allocation policies in a green cloud. A use case is shown in order to illustrate the approach, modeling scattering and saturation allocation policies and comparing them to a purely physical data center scenario. A validation of the proposed model against the CloudSim framework is presented and several numerical results are provided, demonstrating the effectiveness of the approach as a powerful tool for a cloud service provider to perform well-informed decisions about theresource allocation policies to be enforced. © 1990-2012 IEEE.

Cremer L.,College of the Province of Liege | Distefano S.,Polytechnic of Milan | Merlino G.,Messina University | Merlino G.,University of Catania | Puliafito A.,Messina University
Proceedings - 2013 IEEE International Conference on High Performance Computing and Communications, HPCC 2013 and 2013 IEEE International Conference on Embedded and Ubiquitous Computing, EUC 2013 | Year: 2014

Through the Sensing and Actuation as a Service (SAaaS) we aim at implementing a sensing Cloud by enrolling and aggregating sensing resources from sensor networks and personal, mobile devices. We follow a device-driven approach, as adopted in IaaS Clouds: once collected, the physical (sensing) resources are abstracted and virtualised and then provided as a service to end users. A key point of the SAaaS is the abstraction of resources, i.e. providing a uniform way to access to and interact with the underlying physical nodes. In this paper we focus on the sensing resource abstraction in SAaaS, providing details on theoretical and design aspects as well as on technical and implementation ones. In particular, an implementation of the SAaaS abstraction layers for mobiles (SAaaS4mobile) is reported, giving adequate architectural descriptions of the main SAaaS4mobile modules and detailing the actual implementation of such modules for Android-powered mobiles. © 2013 IEEE.

Cremer L.,College of the Province of Liege | Distefano S.,Polytechnic of Milan | Merlino G.,Messina University | Merlino G.,University of Catania | Puliafito A.,Messina University
Communications in Computer and Information Science | Year: 2013

The main goal of the sensing and actuation as a service (SAaaS) approach is to enrol and aggregate sensing resources from heterogeneous sensor networks and smart devices, providing them as a service in a Cloud-wise fashion. SAaaS aims at providing handles on sensing and actuation resources, abstracted and virtualized on top of physical ones, shared by contributors to the SAaaS. This requires adequate mechanisms for letting SAaaS end users interact with the contributing nodes hosting the provided resources. In this paper we focus on such problem, introducing the module of our SAaaS architecture specifically conceived to deal with all the issues related to user-resource interfaces: the Planning Agent (PA). The modular architecture of the PA and its main interactions with SAaaS stakeholders and framework components are described. The development of the PA on the Android platform is detailed, thus implementing a preliminary version of the SAaaS framework, targeted at mobiles: SAaaS4mobile. © Springer-Verlag Berlin Heidelberg 2013.

Discover hidden collaborations