Deep Learning-Driven Compiler Enhancements for Efficient Matrix Multiplication DOI

Raunak Kumar,

Karma Chhering Negi,

Nitish Sharma

et al.

Deleted Journal, Journal Year: 2024, Volume and Issue: 3(2), P. 08 - 18

Published: July 1, 2024

Matrix multiplication is a fundamental operation in many computational fields, requiring optimization to handle increasing data sizes efficiently. In this paper, the implementation of Deep Learning reviewed, which considered important nowadays due growing complexity matrix for gaming and complex programs. The current standard time taken by it on different are described. Tiled multiplication, trims into various pieces calculates product each piece, thereafter combines result, also times both methods were compared. main idea was use Neural Networks (DNN) compare rank code variants that obtained determine their relative performance. A tournament-based ranking system used assigning ranks versions. effectiveness these techniques evaluated operations commonly found deep learning workloads. Up 8.844x speedup over naive size 1024 achieved approach. results demonstrate combining compiler models optimizing multiplication.

Language: Английский

Efficient Text Analysis: A BERT-Based Approach to Named Entity Recognition (NER) and Classification for Malayalam Language DOI

Athira Gopalakrishnan,

K. P. Soman,

Suresh Rajendran

et al.

International Journal of Information Technology, Journal Year: 2025, Volume and Issue: unknown

Published: Jan. 15, 2025

Language: Английский

Citations

0
Extracting microservices from monolithic applications using consistent graph enhanced Graph Transformer DOI

Xianglong Wei,

Jing Li, Xudong He

et al.

Journal of Systems and Software, Journal Year: 2025, Volume and Issue: unknown, P. 112345 - 112345

Published: Jan. 1, 2025

Language: Английский

Citations

0
Designing Microservices Using AI: A Systematic Literature Review DOI Creative Commons
D Devia Narváez, Nicolás Battaglia, Alejandro Fernández

et al.

Software, Journal Year: 2025, Volume and Issue: 4(1), P. 6 - 6

Published: March 19, 2025

Microservices architecture has emerged as a dominant approach for developing scalable and modular software systems, driven by the need agility independent deployability. However, designing these architectures poses significant challenges, particularly in service decomposition, inter-service communication, maintaining data consistency. To address issues, artificial intelligence (AI) techniques, such machine learning (ML) natural language processing (NLP), have been applied with increasing frequency to automate enhance design process. This systematic literature review examines application of AI microservices design, focusing on AI-driven tools methods improving decision-making, architectural validation. analyzes research studies published between 2018 2024 that specifically focus techniques identifying key used, challenges encountered integrating into microservices, emerging trends this area. The findings reveal effectively used optimize performance, tasks, mitigate some complexities inherent architectures. gaps remain areas distributed transactions security. study concludes while offers promising solutions, further empirical is needed refine AI’s role remaining challenges.

Language: Английский

Citations

0
Deep Learning-Driven Compiler Enhancements for Efficient Matrix Multiplication DOI

Raunak Kumar,

Karma Chhering Negi,

Nitish Sharma

et al.

Deleted Journal, Journal Year: 2024, Volume and Issue: 3(2), P. 08 - 18

Published: July 1, 2024

Matrix multiplication is a fundamental operation in many computational fields, requiring optimization to handle increasing data sizes efficiently. In this paper, the implementation of Deep Learning reviewed, which considered important nowadays due growing complexity matrix for gaming and complex programs. The current standard time taken by it on different are described. Tiled multiplication, trims into various pieces calculates product each piece, thereafter combines result, also times both methods were compared. main idea was use Neural Networks (DNN) compare rank code variants that obtained determine their relative performance. A tournament-based ranking system used assigning ranks versions. effectiveness these techniques evaluated operations commonly found deep learning workloads. Up 8.844x speedup over naive size 1024 achieved approach. results demonstrate combining compiler models optimizing multiplication.

Language: Английский

Citations

1