LLM-Based Adaptive Digital Twin Allocation for Microservice Workloads

Pedro Henrique Sachete Garcia, Ester de Souza Oribes, Ivan Mangini Lopes Junior, Braulio Marques de Souza, Angelo Nery Vieira Crestani, Arthur Lorenzon, Marcelo Luizelli, Paulo Silas Severo de Souza, Fábio Rossi

2025

Abstract

Efficient resource allocation in programmable datacenters is a critical challenge due to the diverse and dynamic nature of workloads in cloud-native environments. Traditional methods often fall short in addressing the complexities of modern datacenters, such as inter-service dependencies, latency constraints, and optimal resource utilization. This paper introduces the Dynamic Intelligent Resource Allocation with Large Language Models and Digital Twins (DIRA-LDT) framework, a cutting-edge solution that combines real-time monitoring capabilities of Digital Twins with the predictive and reasoning strengths of Large Language Models (LLMs). DIRA-LDT systematically optimizes resource management by achieving high allocation accuracy, minimizing communication latency, and maximizing bandwidth utilization. By leveraging detailed real-time insights and intelligent decision-making, the framework ensures balanced resource distribution across the datacenter while meeting stringent performance requirements. Among the key results, DIRA-LDT achieves an allocation accuracy of 98.5%, an average latency reduction to 5.3 ms, and a bandwidth utilization of 82.4%, significantly outperforming heuristic-based, statistical, machine learning, and reinforcement learning approaches.

Paper Citation

in Harvard Style

Garcia P., Oribes E., Lopes Junior I., Marques de Souza B., Crestani A., Lorenzon A., Luizelli M., Severo de Souza P. and Rossi F. (2025). LLM-Based Adaptive Digital Twin Allocation for Microservice Workloads. In Proceedings of the 15th International Conference on Cloud Computing and Services Science - Volume 1: CLOSER; ISBN 978-989-758-747-4, SciTePress, pages 61-71. DOI: 10.5220/0013427300003950

in Bibtex Style

@conference{closer25,
author={Pedro Garcia and Ester Oribes and Ivan Lopes Junior and Braulio Marques de Souza and Angelo Crestani and Arthur Lorenzon and Marcelo Luizelli and Paulo Severo de Souza and Fábio Rossi},
title={LLM-Based Adaptive Digital Twin Allocation for Microservice Workloads},
booktitle={Proceedings of the 15th International Conference on Cloud Computing and Services Science - Volume 1: CLOSER},
year={2025},
pages={61-71},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0013427300003950},
isbn={978-989-758-747-4},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 15th International Conference on Cloud Computing and Services Science - Volume 1: CLOSER
TI - LLM-Based Adaptive Digital Twin Allocation for Microservice Workloads
SN - 978-989-758-747-4
AU - Garcia P.
AU - Oribes E.
AU - Lopes Junior I.
AU - Marques de Souza B.
AU - Crestani A.
AU - Lorenzon A.
AU - Luizelli M.
AU - Severo de Souza P.
AU - Rossi F.
PY - 2025
SP - 61
EP - 71
DO - 10.5220/0013427300003950
PB - SciTePress