 |
 |
 |
 |
 |
 |
 |
 |
 |
 |
|
|
 |
|
 |
|
|
|
Secure and Dependable Infrastructure Information infrastructures are becoming central to all of our every day activities. The increasing number of interconnected systems greatly increases the human involvement required for system maintenance and reconfiguration; such concerns lead to the notion of proactive systems (also referred to as autonomic or adaptable systems), in which even complex tuning and maintenance procedures are automated so as to reduce human intervention and to increase overall dependability. Information infrastructures are usually interconnected with critical infrastructures such as those providing power, transportation, and vital human services. Finally, such systems are administrated, operated and used by individuals whose interaction with them might introduce additional sources of failure besides benefiting from human skills. Hence, the dependability of information infrastructures (i) is affected by the dependability of mobile, proactive and embedded systems, and (ii) affects the dependability of other critical infrastructures. contact: Roberto Baldoni Dependable System Design and Integration The integration of system and software components for efficiency, performance and especially dependability requirements is of growing interest from both a research and commercial system design perspective. In this area, our research focuses on the design of fault-tolerant distributed systems using novel replication techniques and standard middleware platforms (e.g., CORBA). In particular, we introduced three-tier architectures for software replication as a viable mean to achieve fault tolerance of a service whose replicas can be deployed on a wide area network. Our replication protocols enforce strong replica consistency and thus give to clients the illusion of interacting with non-replicated services. The IRL system implements the three-tier architecture to increase the availability of CORBA objects. contact: Carlo Marchetti Mobility and Wireless Communication The rapid advance in wireless communication technology and portable computing devices is enabling the realization of mobile computer networks, such as Mobile Ad Hoc Networks (MANETs). The characterizing factor of such a class of networks, i.e. mobility, has a great impact on how the functionalities traditionally allocated at different layers of the ISO/OSI stack are realized, since lower layers cannot easily hide the effects of mobility to the upper ones. In this context, our research focuses on two main topics: (i) to study suitable routing protocols for mobile networking (ii) to address the problem of how applications can interact with lower layers in order to partially dealing with mobility and thus achieving good performance (mobile-aware applications). contact: Roberto Beraldi e-Government and e-Services for cooperative applications The spreading of network technologies and the emergence of Internet allows the development of new interaction business paradigms, commonly referred to as e-Business or e-Commerce. Although this new business paradigm has initially emerged in the commerce context, indeed there are other contexts: specifically, some important initiatives for the definition of what is referred to as e-Government are undertaken in many countries. Modernizing government by using new technologies can be carried out along three main directions:
e-Government systems targeted to electronic service delivery are currently investigated by MIDLAB researchers, as they constitute the basis for supporting efforts in any other direction. In the literature, Cooperative Information Systems (CIS's) are proposed as a possible paradigm for developing systems for e-Government; among the various approaches, the one focused on business process coordination and service-based systems is considered in MIDLAB. Cooperation among different organizations is obtained by sharing and integrating services, commonly referred to as e-Services and Web-Services, across networks. InMIDLAB, the definition and design of an e-Service framework, named PARIDE (Process-based frAmework for oRchestratIon of Dynamic E-services), in which the dynamic integration of different e-Services into cooperative applications is supported, is currently considered. contact:Massimo Mecella Dynamic Networks The research in ubiquitous Internet-size applications require scalable algorithms and architectures for high-troughput information diffusion over large-scale networks. We consider application-level infrastructures with self-reconfiguration capabilities, that can adapt both to underlying network conditions and to application requirements in order to provide a stable performance in changing conditions. The first direction in this study is exploiting dynamic networks in content-based publish/subscribe communication systems. contact: Antonino Virgillito Distribution and optimisation of system resources through Free Software Thetremendous diffusion of ICTs created the issue of a huge quantity of old computers and components to be discarded (E-waste). Its dismantle is not sustainable and this is becoming a global enviromental emergency. Free software allows several degrees of optimization, spannig from OS kernel configuration to fine tuning of applications and services, leading to matching all and only the relevant system and user requirements. Such an approach is used by Trashware movement, aiming to profitably reuse discarded computers as an alternative to dismantling them. Distribution and cooperation of system resources, such as HPC clusters, thin clients, virtual machines, embedded systems, can be effectively combined to Trashware, in order to verify if further optimisation is possible. Objects of such an investigation include but are not limited to computational load, system scalability, (shared) memory process migration, distributed file systems and terminal servers. contact: Davide Lamanna Data-centric communication services Data-centric communications isone of the most interesting abstraction for exchanging information among a set processes which are decoupled in time, space and flow. Distributed shared memory (DSM) is a classic model that provides such data-centric exchanges where decoupling in space, flow and time means that processes can execute wait-free read and write operations on a common replicated variable. The DSM abstraction is implemented by a specific software layer built between the application and the underlying message-passing system. Substantially, this software layer is responsible for maintaining the shared memory consistency. Different consistency models can be supported, e.g. sequential, atomic, causal, FIFO (or PRAM), in order to provide the most appropriate set of abstractions to the application. Each consistency model provides different features in terms of scalability, concurrency, coherence semantics. In particular, causal consistency provides a good trade off respect to the level of semantics coherence, performances and scalability. The research has been focusing on improving existing protocols for ensuring causal consistency providing (1) more efficient protocols that exploit writing semantics, and (2) protocols that allow higher concurrency by relaxing ordering constrains without losing the causal consistency. contact: Sara Tucci Piergiovanni IT Products and Solutions Evaluation The goalof this activity is to evaluate the effectiveness of new technologies, solutions and products that can be used in developing new E-services. contact: Alessandro Termini |