Paper Key : IRJ************195
Author: Sudheer Kandula,Sree Ranga Vasudha Moda
Date Published: 18 Oct 2023
Abstract
In the contemporary landscape of Artificial Intelligence (AI), the convergence of advanced algorithms and massive datasets has ushered in a new era of capabilities and possibilities. At the heart of the modern Artificial Intelligence (AI) workloads lies the significance of data an indispensable resource that fuels the Artificial Intelligence (AI) revolution. Managing, processing, and extracting value at scale from large datasets in runtime demands sophisticated data infrastructure, storage solutions, and computational resources. One foundational component in the High-Performance Computing (HPC), AI infrastructure is Computer Networking, especially in layers such as LLMs (Large Language Models), big-data wrangling and computations. It serves as the backbone that enables efficient data movement, communication, collaboration, and operation of massive workloads with low latencies. A well-designed software system should thrive for optimal utilization of network bandwidth that enhances the performance, speed, accuracy, and scalability of AI systems, allowing organizations to unlock the full potential of Artificial Intelligence (AI). This paper focuses on the data strategies needed to be embedded into our software systems, particularly in the scope of network bandwidth optimization, and outputs a comprehensive comparison on a list of available technologiestechniques for each of the strategies listed.
DOI LINK : 10.56726/IRJMETS45318 https://www.doi.org/10.56726/IRJMETS45318