Carlo Mastroianni

A super-peer model for resource discovery services in large-scale grids

As deployed Grids increase from tens to thousands of nodes, peer-to-peer (P2P) techniques and protocols can be used to implement scalable services and applications. The super-peer model is a novel approach that helps the convergence of P2P models and Grid environments and can be used to deploy a P2P information service in Grids. A super-peer serves a single physical organization in a Grid, and manages metadata associated to the resources provided by the nodes of that organization. Super-peers connect to each other to form a peer network at a higher level. This paper examines how the super-peer model can handle membership management and resource discovery services in a multi-organizational Grid. A simulation analysis evaluates the performance of a resource discovery protocol; simulation results can be used to tune protocol parameters in order to increase search efficiency.

Probabilistic Consolidation of Virtual Machines in Self-Organizing Cloud Data Centers

Power efficiency is one of the main issues that will drive the design of data centers, especially of those devoted to provide Cloud computing services. In virtualized data centers, consolidation of Virtual Machines (VMs) on the minimum number of physical servers has been recognized as a very efficient approach, as this allows unloaded servers to be switched off or used to accommodate more load, which is clearly a cheaper alternative to buy more resources. The consolidation problem must be solved on multiple dimensions, since in modern data centers CPU is not the only critical resource: depending on the characteristics of the workload other resources, for example, RAM and bandwidth, can become the bottleneck. The problem is so complex that centralized and deterministic solutions are practically useless in large data centers with hundreds or thousands of servers. This paper presents ecoCloud, a self-organizing and adaptive approach for the consolidation of VMs on two resources, namely CPU and RAM. Decisions on the assignment and migration of VMs are driven by probabilistic processes and are based exclusively on local information, which makes the approach very simple to implement. Both a fluid-like mathematical model and experiments on a real data center show that the approach rapidly consolidates the workload, and CPU-bound and RAM-bound VMs are balanced, so that both resources are exploited efficiently.

A reversible hierarchical scheme for microcellular systems with overlaying macrocells

Future cellular systems are expected to use multilayered, multisized cells to cover non-homogeneous populated areas. An example in literature is given by a 2 level hierarchical architecture in which an overlaying macrocell provides a group of overflow channels utilized when a microcell, which covers a densely populated area, is not able to accommodate a new call, or a handover from another microcell. The macrocell has the higher hierarchical position, meaning that it can receive handover requests from microcells, lower in the hierarchy, as well as from other macrocells. On the contrary, a call served by the macrocell cannot handover to a microcell. This paper proposes a reversible hierarchical scheme characterized by the presence of handover attempts from macrocells to microcells. The scheme is conceived so that the microcells are given the majority of the traffic load as they are able to operate with very high capacity, while the macrocells, having lower channel utilization, can better carry out their support task. An analytical study is carried out showing that the system performance can be improved, at the expense of relatively little increase of network control overhead, when compared with the classical, i.e. nonreversible hierarchical scheme.

A super-peer model for building resource discovery services in grids: design and simulation analysis

As deployed Grids increase from tens to thousands of nodes, Peer-to-Peer (P2P) techniques and protocols can be used to implement scalable services and applications. The super-peer model is a novel approach that helps the convergence of P2P models and Grid environments and can be used to deploy a P2P information service in Grids. A super-peer serves a single Virtual Organization (VO) in a Grid, and manages metadata associated to the resources provided by the nodes of that VO. Super-peers connect to each other to form a peer network at a higher level. This paper examines how the super-peer model can be used to handle membership management and resource discovery services in a multi-organizational Grid. A simulation analysis evaluates the performance of a resource discovery protocol; simulation results can be used to tune protocol parameters in order to increase search efficiency.

Performance analysis of cellular mobile communication networks supporting multimedia services

This paper illustrates the development of an analytical model for a communication network providing integrated services to a population of mobile users, and presents performance results to both validate the analytical approach, and assess the quality of the services offered to the end users. The analytical model is based on continuous‐time multidimensional birth–death processes, and is focused on just one of the cells in the network. The cellular system is assumed to provide three classes of service: the basic voice service, a data service with bit rate higher than the voice service, and a multimedia service with one voice and one data component. In order to improve the overall network performance, some channels can be reserved to handovers, and multimedia calls that cannot complete a handover are decoupled, by transferring to the target cell only the voice component and suspending the data connection until a sufficient number of channels become free. Numerical results demonstrate the accuracy of the approximate model, as well as the effectiveness of the newly proposed multimedia call decoupling approach.

Designing an information system for Grids: Comparing hierarchical, decentralized P2P and super-peer models

As deployed Grids increase from 10s to 1000s of nodes, the construction of an efficient and scalable information system is a key issue, as it is vital for providing querying and discovery services. Today most Grids adopt a centralized or hierarchical model for their information system, but this model is characterized by poor scalability, resiliency and load-balancing features. Nowadays the research and development community is heading towards the use of scalable information systems based on distributed models such as the decentralized peer-to-peer (P2P) model and the super-peer model. If the former is adopted, each node can act both as a client and a server as it can generate discovery requests and also respond to requests issued by other peers. Requests and responses are forwarded with a hop-by-hop mechanism by ad hoc Grid Services hosted by Grid nodes. The super-peer model is a recently proposed approach that combines features of centralized and P2P models. A super-peer acts as a server for a single Grid organization, and publishes metadata describing the resources provided by the nodes of that organization. At the same time, super-peers connect to each other to form a P2P network at a higher level. This paper analyzes information systems based on three alternative models: the hierarchical, the decentralized P2P, and the super-peer model. A performance evaluation of such models is reported, and afterwards a performance comparison is discussed in order to analyze the pros and cons of each solution.

Self-economy in cloud data centers: statistical assignment and migration of virtual machines

The success of Cloud computing has led to the establishment of large data centers to serve the increasing need for on-demand computational power, but data centers consume a huge amount of electrical power. The problem can be alleviated by mapping virtual machines, VMs, which run client applications, on as few servers as possible, so that some servers with low traffic can be put in low consuming sleep modes. This paper presents a new approach for the adaptive assignment of VMs to servers and their dynamic migration, with a twofold goal: reduce the energy consumption and meet the Service Level Agreements established with users. The approach, based on ant-inspired algorithms, founds on statistical processes: the mapping and migration of VMs are driven by Bernoulli trials whose success probability depends on the utilization of single servers. Experiments highlight the two main advantages with respect to the state of the art: the approach is self-organizing and mostly decentralized, since each server locally decides whether or not a new VM can be served, and the migration process is continuous and adaptive, thus avoiding the need for the simultaneous reassignment of many VMs.

Self-chord: a bio-inspired P2P framework for self-organizing distributed systems

This paper presents “Self-Chord,” a peer-to-peer (P2P) system that inherits the ability of Chord-like structured systems for the construction and maintenance of an overlay of peers, but features enhanced functionalities deriving from ant-inspired algorithms, such as autonomous behavior, self-organization, and capacity to adapt to a changing environment. As opposed to the structured P2P systems deployed so far, resource indexing and placement is uncorrelated with network structure and topology, and resource keys are organized and managed by self-organizing mobile agents through simple local operations driven by probabilistic choices. Self-Chord has three main features that are particularly advantageous in Grid and Cloud Computing: 1) it is possible to give a semantic meaning to keys, which enables the execution of range queries; 2) the keys are fairly distributed over the peers, thus improving the balancing of storage responsibilities; 3) maintenance load is also limited because it is not necessary to reassign keys when new peers or resources are added to the system-the mobile agents will spontaneously reorganize the keys. The efficiency and effectiveness of Self-Chord were assessed both with a simulation framework and with an analytical model inspired by fluid dynamics.

A framework for distributed knowledge management: Design and implementation

This paper describes a framework for implementing distributed ontology-based knowledge management systems (DOKMS). The framework, in particular, focuses on knowledge management within organizations. It investigates the functional requirements to enable Individual Knowledge Workers (IKWs) and distributed communities (e.g., project teams) to create, manage and share knowledge with the support of ontologies. On the one hand, the framework enables distributed and collaborative work by relying on a P2P virtual office model. On the other hand, it provides a multi-layer ontology framework to enable semantics-driven knowledge processing. The ontology framework allows organizational knowledge to be modeled at different levels. An Upper Ontology is exploited to establish a common organizational knowledge background. A set of Workspace Ontologies can be designed to manage, share and search knowledge within communities by the establishment of a contextual (i.e., related to the aim of a group) understanding. Finally, Personal Ontologies support IKWs in personal knowledge management activities. We present an implementation of the designed framework in the K-link+ system and show the suitability of this approach through a use case. The evaluation of K-link+ in a real network is also discussed.

So-Grid: A self-organizing Grid featuring bio-inspired algorithms

This article presents So-Grid, a set of bio-inspired algorithms tailored to the decentralized construction of a Grid information system that features adaptive and self-organization characteristics. Such algorithms exploit the properties of swarm systems, in which a number of entities/agents perform simple operations at the local level, but together engender an advanced form of swarm intelligence at the global level. In particular, So-Grid provides two main functionalities: logical reorganization of resources, inspired by the behavior of some species of ants and termites that move and collect items within their environment, and resource discovery, inspired by the mechanisms through which ants searching for food sources are able to follow the pheromone traces left by other ants. These functionalities are correlated, since an intelligent dissemination can facilitate discovery. In the Grid environment, a number of ant-like agents autonomously travel the Grid through P2P interconnections and use biased probability functions to: (i) replicate resource descriptors in order to favor resource discovery; (ii) collect resource descriptors with similar characteristics in nearby Grid hosts; (iii) foster the dissemination of descriptors corresponding to fresh (recently updated) resources and to resources having high quality of service (QoS) characteristics. Simulation analysis shows that the So-Grid replication algorithm is capable of reducing the entropy of the system and efficiently disseminating content. Moreover, as descriptors are progressively reorganized and replicated, the So-Grid discovery algorithm allows users to reach Grid hosts that store information about a larger number of useful resources in a shorter amount of time. The proposed approach features characteristics, including self-organization, scalability and adaptivity, which make it useful for a dynamic and partially unreliable distributed system.