EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Hardware-Aware Optimization of Large Language Models: A System-Level Analysis

Areej Hamza, Amel Tuama, Asraf Mohamed Moubark

NTU Journal of Engineering and Technology · 28.06.2026 · 5 (2)

DOI: 10.56286/854mxw81 ↗

Kurz gefasst

Fachartikel in der Zeitschrift NTU Journal of Engineering and Technology, online erschienen am 28.06.2026. Verfasst von Areej Hamza und 2 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Deep neural networks (DNNs), particularly large language models (LLMs), have achieved remarkable performance; however, they introduce significant computational and memory challenges that hinder practical deployment. This paper presents a hardware-aware, system-level analysis of key optimization techniques, including pruning, quantization, knowledge distillation, Low-Rank Adaptation (LoRA), and Neural Architecture Search (NAS). The study evaluates these methods in terms of scalability, hardware compatibility, and deployment efficiency. The results show that quantization consistently achieves the highest inference speedups and memory efficiency, while LoRA enables scalable fine-tuning with minimal resource overhead. In contrast, pruning provides limited real-world benefits unless supported by structured sparsity and specialized hardware. These findings highlight the importance of integrated optimization and hardware–algorithm co-design for efficient large-scale AI systems.

Zitieren

APA

Hamza, A., Tuama, A., & Moubark, A. M. (2026). Hardware-Aware Optimization of Large Language Models: A System-Level Analysis. NTU Journal of Engineering and Technology, 5(2). https://doi.org/10.56286/854mxw81

Vancouver

Hamza A, Tuama A, Moubark AM. Hardware-Aware Optimization of Large Language Models: A System-Level Analysis. NTU Journal of Engineering and Technology. 2026;5(2). doi:10.56286/854mxw81

BibTeX
@article{Hamza2026,
  author = {Hamza, Areej and Tuama, Amel and Moubark, Asraf Mohamed},
  title = {Hardware-Aware Optimization of Large Language Models: A System-Level Analysis},
  journal = {NTU Journal of Engineering and Technology},
  year = {2026},
  volume = {5},
  number = {2},
  doi = {10.56286/854mxw81}
}
Kostenlos starten →