Efficient Data Input/Output (I/O) for Finite Difference Time Domain (FDTD). Computation on Graphics Processing Unit (GPU)

Inhaltsverzeichnis (Table of Contents)

• 1 Introduction
• 1.1 Computation in Electromagnetism
• 1.1.1 Maxwell's Equations
• 1.1.2 Finite-Difference Time-Domain (FDTD)
• 1.2 Computational Parallelization Techniques & GPGPU
• 1.2.1 Parallel Computer Architecture
• 1.2.2 Parallel Algorithms & Programs
• 1.2.3 Emerging Parallelization Techniques: GPGPU
• 1.3 The Problem and The Objective
• 1.4 Thesis Overview
• 1.5 Original Contribution
• 2 Electromagnetism & Finite-Difference Time-Domain - Overview
• 2.1 Maxwell's Equations
• 2.2 Finite-Difference Time-Domain (FDTD)
• 2.2.1 Frequency Dependent Material Parameters & Frequency Dependent FDTD
• 2.2.2 Boundary Condition
• 2.3 Summary of Maxwell's Equations and FDTD Method
• 2.4 Computer Implementation of FDTD Method
• 2.4.1 Implementation of FDTD Method
• 2.4.2 Basics of FORTRAN 90 Programming
• 2.5 Advantages and Limitations of FDTD Computation
• 2.6 Concluding Remarks
• 3 Computation of FDTD on GPGPU using CUDA Programming
• 3.1 GPGPU - The Parallelization Monster and Computation Techniques
• 3.2 CUDA and CUDA Fortran
• 3.3 CUDA Implementation of FDTD Method for GPGPU Computation
• 3.4 Computation on Nvidia's General Purpose GPU
• 3.4.1 GPU Hardware and support for FDTD Computation
• 3.4.2 Memory Coalescing
• 3.5 Execution of FDTD Method on GPU Hardware
• 3.6 Concluding Remarks
• 4 The Solution to The Problem
• 4.1 The Problem - Revisited
• 4.2 The Solution
• 4.3 Programmatic Implementation of the Solution
• 4.3.1 Implementation
• 4.3.2 Invoking Buffer Kernel
• 4.4 Possible Limitations and their Solutions
• 4.5 Concluding Remarks
• 5 Evaluation and Validation of The Solution
• 5.1 Testing of the Implemented Solution
• 5.1.1 Input Parameters for FDTD Computation
• 5.1.2 Hardware Environment
• 5.1.3 Test Results
• 5.2 Critical Analysis & Evaluation of Test Results
• 5.2.1 Speed-Up Analysis
• 5.2.2 Evaluation and Comments
• 6 Conclusion and Future Scope
• 6.1 Conclusion
• 6.2 Future Scope

Zielsetzung und Themenschwerpunkte (Objectives and Key Themes)

This dissertation investigates the efficiency of data input/output (I/O) for Finite-Difference Time-Domain (FDTD) computation on Graphics Processing Units (GPUs). The research focuses on optimizing FDTD algorithms for parallel processing on GPU hardware, aiming to enhance computational performance and tackle the challenges associated with data transfer between the CPU and GPU.

• Optimization of FDTD algorithms for parallel processing on GPUs
• Efficient data transfer between CPU and GPU
• Improving computational performance for FDTD simulations
• Exploiting the potential of GPGPU computing for electromagnetic simulations
• Exploring the limitations and solutions for efficient data I/O in FDTD computations on GPUs

Zusammenfassung der Kapitel (Chapter Summaries)

• Chapter 1 provides an introduction to the field of computational electromagnetism, outlining the significance of Maxwell's equations and the FDTD method. It introduces the concept of computational parallelization techniques, particularly GPGPU computing, and highlights the problem addressed in the dissertation.
• Chapter 2 delves into the fundamentals of electromagnetism and the FDTD method, explaining the theoretical basis of the technique and its implementation in computer simulations. It discusses various aspects of the method, including frequency dependent materials, boundary conditions, and implementation using FORTRAN 90.
• Chapter 3 explores the computation of FDTD on GPUs using CUDA programming. It introduces the concept of GPGPU computing, explains CUDA and CUDA Fortran, and details the implementation of the FDTD method on GPU hardware. The chapter also addresses issues like memory coalescing and the execution of FDTD simulations on GPUs.
• Chapter 4 presents the solution to the problem of efficient data I/O for FDTD computations on GPUs. It explains the programmatic implementation of the solution, discusses possible limitations and their solutions, and outlines the effectiveness of the proposed approach.
• Chapter 5 evaluates and validates the implemented solution through testing. It outlines the input parameters, hardware environment, and test results, providing a critical analysis of the results and evaluating the speed-up achieved through the solution.

Schlüsselwörter (Keywords)

This dissertation focuses on the application of GPGPU computing for improving the efficiency of FDTD computations. Key terms and concepts include Finite-Difference Time-Domain (FDTD), Graphics Processing Units (GPUs), CUDA programming, parallel processing, data input/output (I/O), memory coalescing, and computational performance optimization.