Dynamic Reconfiguration Management Based on a Distributed Object Model

Abstract

In this paper we propose a contribution to reduce the complexity of management of dynamic reconfiguration. The model for Dynamic Reconfiguration presented here is based on the object-oriented paradigm and a clear separation between the functionality inside the reconfigured cores and the communication infrastructure. This infrastructure will be the responsible not only for managing data transferences between objects but also for transparently handling other aspects such as state storage and retrieval or controlled object stopping and resuming. INTRODUCTIONMost of the research effort in reconfiguration techniques and design methodologies, is highly technology-dependent [5][6]. Other research efforts have concentrated on higher abstraction levels [7]. However, it is still not clear how to map such high-level services onto current reconfiguration technology. Currently, it is the designer of the reconfigurable application the responsible for solving nontrivial problems, such as structural consistency, state persistence, etcThe solution we propose in this paper is based on the object-oriented paradigm, instead of the task-based approach that has been normally used as the programming model for reconfigurable applications. Object technology is better suited to nowadays reconfiguration technology since it enforces a systematic way to deal with state, execution of operations and encapsulation. The definition of a clear object interface, and the use of the distributed objects communication model, properly extended to support dynamic reconfiguration, will provide the necessary orthogonalization between object functionality and communication-reconfiguration tasks.

Membrane-based design and management methodology for parallel dynamically reconfigurable embedded systems

“…This is exactly what we want to do in the hardware world. After the transition, of software designers, from task-based models to the object-based ones [21], moving to componentbased models seems to be an obvious solution for the future. In fact, such models allow for the abstraction level and the granularity we are expecting.…”

A Reasoning Hardware Platform for Real-Time Common-Sense Inference

“…New entries can be added to the existing SNs through writing operations performed on local memories. The work proposed here resorts to the infrastructure for dynamic reconfiguration management for Xilinx platforms developed by Dondo et al [ 22 ].…”