Tehnički vjesnik, Vol. 32 No. 1, 2025.
Izvorni znanstveni članak
https://doi.org/10.17559/TV-20240521001625
Enhancing Cloud Resource Allocation with TrustFusionNet Using Random Forests and Convolutional Neural Networks
S. T. Bharathi
; Department of Computer Science and Engineering, Kalasalingam Academy of Research and Academy, Srivilliputhur, Tamilnadu, India
*
C. Balasubramanian
; Department of Computer Science and Engineering, Kalasalingam Academy of Research and Academy, Srivilliputhur, Tamilnadu, India
S. Shanmugapriya
; Department of Computer Science and Engineering, Kalasalingam Academy of Research and Academy, Srivilliputhur, Tamilnadu, India
* Dopisni autor.
Sažetak
In the cloud computing, trust awareness and efficient resource allocation are central concerns. This research paper introduces an innovative resource allocation framework, "TrustFusionNet," which combines two prolific algorithms: Random Forest and Convolutional Neural Networks (CNNs), to comprehensively assess the trustworthiness of cloud resources. TrustFusionNet systematically evaluates resource trust values, taking into account diverse resource attributes, historical data, and behavioral patterns. These trust values are then integrated with trust-enhanced features extracted by CNNs, yielding a holistic trust assessment. The need for trust awareness in cloud computing arises from the imperative to ensure that cloud resources are dependable and secure. TrustFusionNet addresses this challenge by providing a multifaceted trust assessment process. Random Forest, renowned for its ensemble learning capabilities, aids in interpreting trust scores based on various resource factors. Meanwhile, CNNs excel in extracting intricate trust-related features from resource data, capturing subtle nuances that conventional methods may overlook. To rigorously evaluate the efficacy of TrustFusionNet, extensive simulation analyses are conducted. Performance comparisons are made against established resource allocation algorithms, employing a comprehensive set of simulation metrics. These metrics encompass resource utilization, trust assurance, allocation efficiency, and system stability. The findings reveal that TrustFusionNet surpasses existing algorithms in enhancing trust assurance and optimizing resource allocation in the cloud computing domain. This research paper paves the way for the advancement of resource allocation using trust awareness in cloud computing, emphasizing the importance of trust-aware decision-making. TrustFusionNet exemplifies a promising approach that balances interpretability and deep feature extraction, promising robust and secure resource allocation. By addressing the paramount issue of trust assurance, it paves the way for more dependable cloud computing ecosystems.
Ključne riječi
cloud computing; convolutional neural networks; resource allocation; trust awareness; TrustFusionNet
Hrčak ID:
325854
URI
Datum izdavanja:
31.12.2024.
Posjeta: 7 *