1,721,104 research outputs found
A Modeling Framework for Unstructured Supernode Networks
No full textEvery peer-to-peer system is based on an overlay scheme, which defines how peers are connected, how messages are propagated among nodes to share resources and information, and what security mechanisms are adopted. In this letter, we propose a modeling framework for unstructured 2-layered overlay schemes, also known as Unstructured Supernode Networks (USNs). Moreover, we present an application of the proposed methodology, i.e. search cost analysis
Review of Elements of Parallel Computing
Full text linkAs the title clearly states, this book is about parallel computing. Modern computers are no longer characterized by a single, fully sequential CPU. Instead, they have one or more multicore/manycore processors. The purpose of such parallel architectures is to enable the simultaneous execution of instructions, in order to achieve faster computations. In high performance computing, clusters of parallel processors are used to achieve PFLOPS performance, which is necessary for scientific and Big Data applications.
Mastering parallel computing means having deep knowledge of parallel architectures, parallel programming models, parallel algorithms, parallel design patterns, and performance analysis and optimization techniques. The design of parallel programs requires a lot of creativity, because there is no universal recipe that allows one to achieve the best possible efficiency for any problem.
The book presents the fundamental concepts of parallel computing from the point of view of the algorithmic and implementation patterns. The idea is that, while the hardware keeps changing, the same principles of parallel computing are reused. The book surveys some key algorithmic structures and programming models, together with an abstract representation of the underlying hardware. Parallel programming patterns are purposely not illustrated using the formal design patterns approach, to keep an informal and friendly presentation that is suited to novices
Introducing artificial evolution into peer-to-peer networks with the distributed remodeling framework
No full textA peer-to-peer (P2P) network is a complex system whose elements (peer nodes, or simply peers) cooperate to implement scalable distributed services. From a general point of view, the activities of a P2P system are consequences of external inputs coming from the environment, and of the internal feedbacks among nodes. The reaction of a peer to direct or indirect inputs from the environment is dictated by its functional structure, which is usually defined in terms of static rules (protocols) shared among peers. The introduction of artificial evolution mechanisms may improve the efficiency of P2P networks, with respect to resource consumption, while preserving high performance in response to the environmental needs. In this paper, we propose the distributed remodeling framework (DRF), a general approach for the design of efficient environment-driven peer-to-peer networks. As a case study, we consider an ultra-large-scale storage and computing system whose nodes perform lookups for resources provided by other nodes, to cope with task execution requests that cannot be fulfilled locally. Thanks to the DRF, workload modifications trigger reconfigurations at the level of single peers, from which global system adaptation emerges without centralized control
Evolutionary strategies for ultra-large-scale autonomic systems
No full textUltra-large-scale (ULS) systems originate from the need to address social problems that are getting more and more complex, such as climatic monitoring, transportation, citizens protection and security. These factors imply a continuous increase of information systems, requiring digital communication networks allowing for communication between people, between machines, and machines and people. The aim of this paper is to present a novel approach for the design of highly adaptive ULS systems, with the focus on computer-supported evolution, adaptable structures, emergent behaviors as well as advanced monitoring and control techniques. We illustrate the Networked Autonomic Machine (NAM), a framework for the characterization of self-managing, highly reconfigurable ULS systems, and the Adaptive Evolutionary Framework (AEF), for the implementation of distributed evolutionary strategies. Finally, we show their effectiveness in the design of an ULS overlay network provided with evolutionary strategies for facing denial-of-service attacks
The Peer-to-Peer Paradigm Applied to Hydrogen Energy Distribution
No full textThe shift from centralized power plants to fully distributed generation, storage and provision of energy may be the greatest innovation of next decades. To succeed, it must be supported by the completion of the decarbonization process, using hydrogen as main energy carrier. The envisioned new energy economy (hydrogen economy) promises to eliminate global warming, pollution and dependency on oil reserves. Among the many issues that still have to be solved, in this paper we focus on the ICT aspect. Distributed energy production and sharing requires efficient distributed software architectures. To this purpose, the peer-to-peer paradigm is highly suitable for the realization of robust and scalable systems
Towards fully autonomic peer-to-peer systems
No full textAbstractLarge-scale distributed applications are becoming more and more demanding in terms of efficiency and flexibility of the technological infrastructure, for which traditional solutions based on the client/server paradigm are not suitable. The peer-to-peer paradigm provides an appealing solution to this problem, allowing to deploy robust networks of collectors, providers and consumers of resources. One step beyond, in an autonomic computing vision, the structure of each peer, or at least its configuration, may be dynamically adjusted by an adaptive plan which determines successive re-configurations in response to the environment, and turns the P2P network in a complex adaptive system (CAS). We analyze a Grid Computing architecture whose resource sharing mechanisms are based on this framework
Towards a Peer-to-Peer Hydrogen Economy Framework
No full textThe research community is seeking for novel technological/business models to speed up the decarbonization process, i.e. a decreasing relative reliance on carbon. The completion of decarbonization ultimately depends on the production and use of pure hydrogen as energy carrier. With respect to electricity, which is currently the most relevant and clean energy carrier, hydrogen has a fundamental advantage that it can be stored efficiently. In this research context, we propose a Peer-to-Peer Hydrogen Economy Framework based on decentralized production, storage and trading of energy. We apply the peer-to-peer paradigm for designing a virtual network, where peer software entities implement distributed algorithms for the localization of remote energy providers. The proposed approach has many advantages, both technical (availability, robustness, scalability) and socio-economic (shared responsibilities and improved competition)
A Design Framework for Ultra-Large-Scale Autonomic Systems
No full textThe origins of ultra-large-scale (ULS) systems derive from social problems that are getting more and more complex, such as climatic monitoring, transportation, citizens protection and security. These factors imply a continuous increase of information systems that evolve towards ultra-dimension systems, requiring digital communication networks that allow for communication between people, between objects, and objects and people. The aim of this paper is to present novel approaches for the engineering of highly adaptive ULS systems, with the focus on computer-supported evolution, adaptable structure, emergent behaviors as well as advanced monitoring and control techniques. We illustrate the Networked Autonomic Machine (NAM), a framework for the characterization of the elements of self-*, highly dynamic ULS systems. Moreover, we recall the Adaptive Evolutionary Framework (AEF), for the implementation of distributed evolutionary strategies. Finally, we describe an example scenario of large peer-to-peer network under targeted attacks, showing the benefits of the NAM-AEF design
Distributed reputation management for service-oriented peer-to-peer enterprise communities
No full textPeer-to-peer service sharing might allow for unmediated, almost instantaneous inter-firm collaboration, yet robust reputation management techniques are strongly required to withstand fraudulent actions of malicious peers. In this paper we illustrate a distributed reputation management system for service-oriented peer-to-peer networks, called SAFE, which exploits voting and eectively copes with trust misrepresentation attempts. Simulation results show that SAFE allows peers to select high-quality services, even when a large extent of malicious providers and advisors are active in the community
- …
