Content area
Abstract
Graph analytics has become essential for extracting insights from complex, interconnected data across diverse domains such as social networks, biological systems, and natural language processing. However, as data volume and complexity continue to grow, traditional graph processing techniques face significant scalability challenges, limiting their efficiency and effectiveness. Concurrently, advancements in high-performance computing (HPC) and machine learning (ML) offer promising solutions to address these limitations by enhancing computational efficiency and analytical depth.
This dissertation investigates the fundamental challenges in scaling large-scale graph analytics and explores how the integration of HPC and ML can lead to more efficient, scalable, and adaptive analytical frameworks. Specifically, we examine how HPC infrastructure can improve the performance of graph processing algorithms, how ML models can be leveraged to address inherent limitations in traditional graph analytics, and how advancements in graph analytics can, in turn, refine and enhance machine learning techniques. By bridging the gap between these fields, this research aims to contribute to the development of next-generation, high-performance graph analytics frameworks capable of handling dynamic, large-scale datasets with greater efficiency and adaptability.
Details
