Vol 5, No 2 (2021)

Parallel and Distributed Numerical Algorithms: Concepts, Methods, and Applications

Authors: Savrav Dubey, Deepali Mehta

Abstract: The rapid growth of scientific computing and data-driven engineering applications has led to the increasing demand for solving large-scale numerical problems efficiently. Traditional sequential numerical algorithms often fail to meet performance and memory requirements when applied to high-dimensional or large systems of equations. Parallel and distributed numerical algorithms address these challenges by exploiting multiple processors and distributed memory systems to reduce computation time and enhance scalability. This paper presents a comprehensive review of parallel and distributed numerical algorithms, focusing on their theoretical foundations, algorithmic structures, communication models, and practical implementation issues. Key numerical problems such as linear systems, eigenvalue computations, numerical integration, and partial differential equations are discussed in the context of parallelization strategies. The paper also highlights performance metrics, programming models, and current challenges associated with distributed computing environments. The discussion aims to provide a balanced understanding of both classical approaches and recent developments in parallel numerical computing.

Keywords: Parallel computing, distributed algorithms, numerical methods, high performance computing, scalability, scientific computing

Full Issue

View or download the full issue PDF 77-88

Table of Contents