Airam Jonatas Preto

Processing Mesoscale Climatology in a Grid Environment

Enhancing the quality of weather and climate forecasts are central scientific research objectives worldwide. However, simulations of the atmosphere, usually demand high processing power and large storage resources. In this context, we present the GBRAMS project, that applies grid computing to speed up the generation of a regional model climatology for Brazil. A grid infrastructure was built to perform long-term integrations of a mesoscale numerical model (BRAMS), managing a queue of up to nine independent jobs submitted to three clusters spread over Brazil- Three distinct middlewares, Globus Toolkit, OurGrid and OAR/CIGRI, were compared in their ability to manage these jobs, and results on the usage of each node of the grid are provided. We analyze the impact of the resulted climatology in the accuracy of climate forecast, showing model bias removal which indicates correctness of the generated climatology. Our central contribution are how to use grid computing to speed-up climatology generation and the middleware impact on this enterprise.

Interpolating EFGM for computing continuous and discontinuous electromagnetic fields

Purpose – This paper proposes an interpolating approach of the element‐free Galerkin method (EFGM) coupled with a modified truncation scheme for solving Poissons boundary value problems in domains involving material non‐homogeneities. The suitability and efficiency of the proposed implementation are evaluated for a given set of test cases of electrostatic field in domains involving different material interfaces.Design/methodology/approach – The authors combined an interpolating approximation with a modified domain truncation scheme, which avoids additional techniques for enforcing the Dirichlet boundary conditions and for dealing with material interfaces usually employed in meshfree formulations.Findings – The local electric potential and field distributions were correctly described as well as the global quantities like the total potency and resistance. Since, the treatment of the material interfaces becomes practically the same for both the finite element method (FEM) and the proposed EFGM, FEM‐oriented...

A parallel implementation of the LTSn method for a radiative transfer problem

A radiative transfer solver that implements the LTSn method was optimized and parallelized using the MPI message passing communication library. Timing and profiling information was obtained for the sequential code in order to identify performance bottlenecks. Performance tests were executed in a distributed memory parallel machine, a multicomputer based on IA-32 architecture. The radiative transfer equation was solved for a cloud test case to evaluate the parallel performance of the LTSn method. The LTSn code includes spatial discretization of the domain and Fourier decomposition of the radiances leading to independent azimuthal modes. This yields an independent radiative transfer equation for each mode that can be executed by a different processor in a parallel implementation. Speed-up results show that the parallel implementation is suitable for the used architecture.

A parallel engine for graphical interactive molecular dynamics simulations

The current work proposes a parallel implementation for interactive molecular dynamics simulations (MD). The interactive capability is modeled by finite automata that are executed in the processing nodes. Any interaction implies in a communication between the user interface and the finite automata. The ADKS, an interactive sequential MD code that provides graphical output was chosen as a case study. A parallel version of this code was developed using the MPI communication library to check its parallel performance without/with visualization. Performance results are discussed for both cases and the influence of visualization in the performance is also treated, including image update rate. In order to allow a modular approach, a new parallel version of the ADKS is being implemented employing the PyMPI Python extension.

A neural network implementation for data assimilation using MPI

The meteorological data assimilation process can be described as a procedure that uses observational data to improve the weather forecast produced by means of a mathematical model. Traditional metlods include deterministic and probabilistic approaches. Included in the latter class, Kahnan filtering has the advantage of minimizing tie error propagated during the data assimilation. However, this method demands a heavy computational power as it usually deals with meteorological parameters for a huge number of grid points. Recently, neural networks have been proposed as a new method for meteorological data assimilation by employing a Multilayer Perception network with backpropagation learning to emulate Kalrnan filtering at a lower computational cost. In order to further reduce that cost and to overcome performance bottlenecks of the training phase, code opdmizations were performe& making use of classical, hardware-independent techniques and Fortran 95 intrinsic functions, thus eliminating inefficient routines. A former attempt to parallelize the code and run it in a 4-processor shared memory machine, made use of HPF (High Performance Fortran) dwectives imbedded in the optimized code. Performance was poor and therefore a new version of the Fortran 95 code was developed using Message Passing Interface library calls. This code was executed in a 16-node multicomputer showing good speed-up up to 4 processors.

Design of Electrooptic Modulators Using a Multiobjective Optimization Approach

This paper presents preliminary results of a multiobjective approach for optimization of design parameters of Mach-Zehnder-based lithium niobate modulators. This process uses a genetic algorithm to iteratively refine candidate sets of parameters, and the characterization of the modulators is performed by the finite-element method. Test cases include optimization of a conventional Mach-Zehnder modulator and a variation of this kind of device using additional floating electrodes. The set of characteristics includes parameters such as the characteristic impedance, the effective index of the microwave, and the half-wave voltage.

Challenges for mesoscale climatology execution on experimental grid computing systems

This paper discusses the challenges of executing a long-term application on a computational grid, which generates the climatology of the atmospheric numerical model BRAMS (Brazilian development on Regional Atmospheric Modeling System) using ensemble members. We have developed a workflow that submits climatology to the computational grid composed by three different grid middlewares (OurGrid, OAR/CiGri and Globus) and three clusters (situated in Porto Alegre, São José dos Campos and Cachoeira Paulista—Brazil). The application characteristics demand a processing grid, rather than a data grid, due to intensive computation and data transfer between the geographically distributed grid nodes. We achieved the goal of generating the climatology using a computational grid. However, we observed problems on application performance due data transfer and non-availability of the computational grid. Questions related to data storage/transfer and grid failures must be better treated to ensure application performance.

A multiobjective approach for optimizing electrooptic modulators

This work presents preliminary results of a multiobjective approach for optimization of some design parameters of Mach-Zehnder-based lithium niobate modulators. This process uses a genetic algorithm to iteratively refine candidate sets of parameters and the characterization of the modulators is performed by the Finite Element Method. Test cases include optimization of a conventional Mach-Zehnder modulator and a variation of this kind of device using additional floating electrodes. The set of characteristics include parameters such as Zc, Nm and VpiL.

A new parallel environment for interactive simulations implementing safe multithreading with MPI

This work presents a new parallel environment for interactive simulations. This environment integrates a MPI-based parallel simulation engine, a visualization module, and a user interface that supports modification of simulation parameters and visualization at runtime. This requires multiple threads, one to execute the simulation or the visualization, and other to receive user input. Since many MPI implementations are not thread-safe, it is proposed a new parallel extension of the Python environment that uses UDP sockets in addition to the calls to the MPI library functions. This approach preserves interactivity, which is required to allow researchers to modify simulation parameters and to visualize results at runtime. The ADKS simulator was chosen as a case study. It is a sequential interactive software for molecular dynamics simulations used in the study of defects in solid materials. The simulation engine was parallelized using non-blocking communication and speedups very close to linear were obtained in the test cases. The proposed approach can be extended to be employed in high performance distributed computing.