António Esteves

An Evolutionary Approach to the Use of Petri Net Based Models

The main purpose of this article is to present how Petri Nets (PNs) have been used for hardware design at our research laboratory. We describe the use of PN models to specify synchronous parallel controllers and how PN specifications can be extended to include the behavioural description of the data path, by using object-oriented concepts. Some hierarchical mechanisms which deal with the specification of complex digital systems are highlighted. It is described a design flow that includes, among others, the automatic generation of VHDL code to synthesize the control unit of the system. The use of PNs as part of a multiple-view model within an object-oriented methodology for hardware/software codesign is debated. The EDgAR-2 platform is considered as the reconfigurable target architecture for implementing the systems and its main characteristics are shown.

Simulation of the nucleation of the precipitate Al 3 Sc in an Aluminum Scandium alloy using the kinetic Monte Carlo method

This paper describes the simulation of the phenomenon of nucleation of the precipitate Al3Sc in an Aluminum Scandium alloy using the kinetic Monte Carlo (kMC) method and the density-based clustering with noise (DBSCAN) method to filter the simulation data. To conduct this task, kMC and DBSCAN algorithms were implemented in C language. The study covers a range of temperatures, concentrations, and dimensions, going from 573K to 873K, 0.25% to 5%, and 50×50×50 to 100×100×100. The Al3Sc precipitation was successfully simulated at the atomistic scale. DBSCAN revealed to be a valorous aid to identify the precipitates. The achieved results are in good agreement with those reported in the literature, but we went deeper in the evaluation of the influence of all the simulation and analysis parameters.

Distributed Memory Implementation Strategies for the kinetic Monte Carlo Algorithm

This paper presents strategies to parallelize a previously implemented kinetic Monte Carlo (kMC) algorithm. The process under simulation is the precipitation in an aluminum scandium alloy. The selected parallel algorithm is called synchronous parallel kinetic Monte Carlo (spkMC). spkMC was implemented with a distributed memory architecture and using the Message Passing Interface (MPI) communication protocol. In spkMC the different processes synchronize at regular points, called end of sprint. During a sprint there is no interaction among processes. A checker board scheme was adopted to avoid possible conflicts among processes during each sprint. To optimize performance different implementations were explored, each one with a different computation vs. communication strategy. The obtained results prove that a rigorous distributed and parallel implementation reproduces accurately the statistical behavior observed with the sequential kMC. Results also prove that simulation time can be reduced with a distributed parallelization but, due to the non-deterministic nature of kMC, significant and scalable gains in parallelization oblige to introduce some simplifications and approximations.

EDgAR: A Platform for Hardware/Software Codesign

Codesign is a unified methodology to develop complex systems with hardware and software components. EDgAR, a platform for hardware/software codesign is described, which is intended to prototype complex digital systems. It employs programmable logic devices (MACHs and FPGAs) and a transputer-based parallel architecture. This platform and its associated methodology reduce the systems production cost, decreasing the time for the design and the test of the prototypes. The EDgAR supporting tools are introduced, which were conceived to specify systems at a high-level of abstraction, with a standard language and to allow a high degree of automation on the synthesis process. This platform was used to emulate an integrated circuit for image processing purposes.