Vai al contenuto| Home page|

   Ti trovi in: HOME »Programmi, progetti e risultati »I progetti »PRIN - Programmi di ricerca di Rilevante Interesse Nazionale»Programma di ricerca
INIZIO_TESTO_DA_INDICIZZARE

RESEARCH PROGRAM

italiano - inglese

COMMUTA: Mutant hardware/software components for dynamically reconfigurable distributed systems

Politecnico di Torino
Abstract
The COMMUTA project aims at defining, constructing and experimenting a distributed reconfigurable architecture based on the possibility of dynamically modifying the hardware and software components of a system.

The most innovative aspect of this project is the ability to handle both hardware and software reconfigurability using Mobile Agents.

Mobile agents are software programs that can be dispatched from one system to another to be executed in a remote environment. The COMMUTA project will use mobile agents to enforce requirements, constraints, standards, configurations and procedures, which are specific in a particular system, but can be unknown for the producers of some individuals that participate in the system itself.

The COMMUTA project will push the concept of Mobile Agents to the hardware level introducing the concept of hardware mobility. The use of uncommitted logic, whose function can be modified at runtime to optimize performances or to replace a faulty hardware component, is very appealing. Hardware mobility has never been applied to self-healing systems where we believe it has the potential of becoming a significant breakthrough.

The functionality of the architecture will be experimented in a real application for the control of distributed sensors. This application is particularly important and requires controlling several distributed devices in wide geographic areas. In this case, we assume that the >>>

Principal Investigator
Paolo PRINETTO Politecnico di TORINO
Research Objectives
The ever-growing evolution of distributed systems enables several application scenarios, where a large variety of uses of hardware/software components exists. This is especially true of distributed systems used in industrial control applications, including environmental control and mobile devices control. In all such scenarios it is necessary to consider requirements related to the distributed nature of components, their localization, the need for automatic mechanisms enabling system reconfiguration, relying when possible on the network of the distributed system itself, and for integration of heterogeneous systems which often have limited hardware/software resources due to physical constraints (e.g., limited power consumption).
Such characteristics are particularly important for applications which need to use numerous computing nodes distributed over large areas, often heterogeneous and hard to reach. In addition, the nature of industrial systems in itself entails the need for the design of reliable, secure, and high-performance systems in many applications.
In this context, challenging open-issues for both research and applications include:
• managing secure network access
• localizing of mobile units
• providing high-performance in order to both use complex techniques of distributed computing, and exchange data in a secure and efficient way (including compression and cryptographic techniques)
• providing support for profiles and >>>

Timescale
24 months
National and international background
Today, the use of digital systems pervades all areas of our lives. The trend of producing digitally-aware environments keeps growing, and it is very likely that in the future we will be surrounded by many heterogeneous devices, from domestic systems to complex applications like automotive, transportations, and medical control systems, which communicate each other through distributed and wireless interfaces. We will see a new society of digital systems "living and evolving" around and within our society. The individuals of this new society will be ubiquitous and heterogeneous digital systems that will provide high productivity and great flexibility. The communication infrastructure of this society will rely almost entirely on wireless connections.

Pervasive and ubiquitous distributed systems are the reference scenario of the COMMUTA project. Besides gaining a vital role in digitally-aware environments, pervasive and ubiquitous distributed systems also rise a number of challenging requirements that could turn out to limit the actual realization of such scenarios. These include the need for scalability, customizability of services, flexibility, and extensibility [FU98].

There have been many attempts to provide effective answers to such needs. Most of the proposed approaches, however, try to adapt well-established models and technologies within the new setting, and usually take for granted the traditional client-server architecture. For example >>>