International audienceThrough the recent emergence of joint resource and network virtualization, dynamic composition and provisioning of time-limited and isolated virtual infrastructures is now possible. One other benefit of infrastructure virtualization is the capability of transparent reliability provisioning (reliability becomes a service provided by the infrastructure). In this context, we discuss the motivations and gains of introducing customizable reliability of virtual infrastructures when executing large-scale distributed applications, and present a framework to specify, allocate and deploy virtualized infrastructure with reliability capabilities. An approach to efficiently specify and control the reliability at runtime is proposed. We illustrate these ideas by analyzing the introduction of reliability at the virtual-infrastructure level on a real application. Experimental results, obtained with an actual medical-imaging application running in virtual infrastructures provisioned in the experimental large-scale Grid'5000 platform, show the benefits of the virtualization of reliability
Abstract. Over the years, the Internet has become a central tool for society. The extent of its growth and usage raises critical issues associated with its design principles that need to be addressed before it reaches its limits. Many emerging applications have increasing requirements in terms of bandwidth, QoS and manageability. Moreover, applications such as Cloud computing and 3D-video streaming require optimization and combined provisioning of different infrastructure resources and services that include both network and IT resources. Demands become more and more sporadic and variable, making dynamic provisioning highly needed. As a huge energy consumer, the Internet also needs to be energyconscious. Applications critical for society and business (e.g., health, finance) or for real-time communication demand a highly reliable, robust and secure Internet. Finally, the future Internet needs to support sustainable business models, in order to drive innovation, competition, and research. Combining optical network technology with Cloud technology is key to addressing the future Internet/Cloud challenges. In this con- text, we propose an integrated approach: realizing the convergence of the IT-and optical-network-provisioning models will help bring revenues to all the actors involved in the value chain. Premium advanced network and IT managed services integrated with the vanilla Internet will ensure a sustainable future Internet/Cloud enabling demanding and ubiquitous applications to coexist.
ABSTRACT:The European Integrated Project GEYSERS -Generalised Architecture for Dynamic Infrastructure Services -is concentrating on infrastructures incorporating integrated optical network and IT resources in support of the Future Internet with special emphasis on cloud computing. More specifically GEYSERS proposes the concept of Virtual Infrastructures over one or more interconnected Physical Infrastructures comprising both network and IT resources. Taking into consideration the energy consumption levels associated with the ICT today and the expansion of the Internet in size and complexity, that incurring increased energy consumption of both IT and network resources, energy efficient infrastructure design becomes critical. To address this need, in the framework of GEYSERS, we propose energy efficient design of infrastructures incorporating integrated optical network and IT resources, supporting resilient end-to-end services. Our modeling results quantify significant energy savings of the proposed solution by jointly optimizing the allocation of both network and IT resources.
SUMMARYGrid applications move large amounts of data between distributed resources, and the efficiency of a Grid depends on their timely delivery within given bounds (deadlines). In most cases, the data volume and deadline are known in advance, allowing for both network planning and connection admission control (CAC). We formally define the problem and, based on this formalization, describe the operation of a feasible procedure for network reservations of deadline-constrained bulk data transfer requests. The procedure guarantees a minimum bandwidth to meet the deadlines and allows for opportunistic utilization of residual network capacity. We propose a novel analytical model based on the solution of an M/M(nc)/1/k(s)-R P S queue. The analytical model is validated against ns-2 simulations taking into account network level details (IP and TCP protocols), showing remarkably good coherence even under heavy loads. The model is orders of magnitude faster than simulation, which enables its application to plan the capacity of Grid networks, and to enforce CAC under the hypothesis of a dominating bottleneck on the transfer route.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.