Institutul de Matematică şi Informatică "Vladimir Andrunachievici"
Universitatea de Stat din MoldovaRO EN
Efficient GPU Power Management through Advanced Framework Utilizing Optimization Algorithms
Authors: Ramesha Rehman, Mashood Ul Haq Chishti, Hamza Yamin
Keywords: GPU, power reduction, machine learning, Particle Swarm Optimization.
– 0.97 Mb
Keywords: GPU, power reduction, machine learning, Particle Swarm Optimization.
Abstract
The rapid rise in power usage by GPUs due to advances in machine and deep learning has led to an increase in power consumption of GPUs in Deep Learning workloads. To address this issue, a novel research project focuses on integrating Particle Swarm Optimization into a model training optimization framework to effectively reduce GPU power consumption during machine learning and deep learning training workloads. By utilizing the Particle Swarm Optimization (PSO)\protect\hyperlink{b1}{{[}1{]}} algorithm within the proposed framework, we show the effectiveness of PSO in creating a more efficient power management strategy while also maintaining the performance. Upon evaluation of the proposed framework, it shows a reduction of 15.8\% to 75.8\% in power consumption across multiple workloads, with little to no performance loss.1;2;3Lahore Garrison University
Phase 6, Block C, Defence Housing Authority,
Lahore, Punjab, Pakistan 54810
Ramesha Rehman
ORCID: https://orcid.org/0009-0009-3322-117X
E-mail:
Mashood Ul Haq Chishti
ORCID: https://orcid.org/0009-0000-4024-2360
E-mail:
Hamza Yamin
ORCID: https://orcid.org/0009-0008-9556-2730
E-mail:
Phase 6, Block C, Defence Housing Authority,
Lahore, Punjab, Pakistan 54810
Ramesha Rehman
ORCID: https://orcid.org/0009-0009-3322-117X
E-mail:
Mashood Ul Haq Chishti
ORCID: https://orcid.org/0009-0000-4024-2360
E-mail:
Hamza Yamin
ORCID: https://orcid.org/0009-0008-9556-2730
E-mail:
DOI
https://doi.org/10.56415/csjm.v32.08Fulltext
– 0.97 Mb
Contents
- Formal Analysis of Medical Systems using Multi-Agent Systems with Information Sharing
- Outer independent total double Italian domination number
- On the trees with maximum Cardinality-Redundance number
- Vector finite fields of characteristic two as algebraic support of multivariate cryptography
- Optimizing Cervical Cancer Classification with SVM and Improved Genetic Algorithm on Pap Smear Images
- A Binary Grey Wolf Optimizer with Mutation for Mining Association Rules
- A Coloured Petri Net-based approach and Genetic Algorithms for improving services in the Emergency Department
- Efficient GPU Power Management through Advanced Framework Utilizing Optimization Algorithms
