Cloud & High-Performance Computing Laboratory
The paper “Optimizing Global Federated Learning: A Serverless Hierarchical Approach with Region-Aware Placement” has been accepted for publication and presentation at the 16th International Conference on Cloud Computing and Services Science (CLOSER 2026).
The event will be held in Benidorm, Spain, from May 19 to 21, 2026.
The work was authored by student Matheus Marotti (UFF) and Professor Lúcia Drummond (UFF).
Miguel de Lima, a master’s student from UFF advised by Lúcia Drummond, visited TADaaM (Inria Bordeaux) from November 25 to December 8, 2025, as part of the DecoHPC project.
During his visit, Miguel presented the paper “Towards a Novel Vertical Scaling Approach for Bursty Workloads in Kubernetes” at CLOUDAM, a workshop co-located with the IEEE/ACM International Conference on Utility and Cloud Computing (UCC 2025), held in Nantes, France, from December 1 to 4, 2025.
The paper is authored by Miguel Lima (UFF), Lúcia Drummond (UFF), and Luan Teylo (INRIA), focusing on containerization and resource management strategies for HPC workloads.
The paper “Análise de um Log Real de Jobs visando a Previsão de Tempos de Espera para Execução em um Cluster de Alto Desempenho (Analysis of a Real Job Log for Predicting Waiting Times for Execution in a High-Performance Cluster)” was presented in the Post-Graduation Forum at the X Southeast Regional School of High Performance (ERAD-SE 2025).
The event was hosted by the Institute of Computing at the Universidade Federal Fluminense (UFF), in Niterói, Brazil, from November 5-7, 2025.
The work was authored by Bernardo Gallo (UFF), Matheus Marotti (UFF), Lúcia Maria de Assumpção Drummond (UFF), José Viterbo (UFF), Felipe A. Portella (PETROBRAS), Paulo J. B. Estrela (PETROBRAS), and Renzo Q. Malini (PETROBRAS).
The paper “Explorando Modelos Preditivos de Tempo de Execução de Simulações de Reservatório em Clusters HPC (Exploring Predictive Models for Execution Time of Reservoir Simulations in HPC Clusters)” was recognized with the Best Paper Award in the Post-Graduation Forum at the X Southeast Regional School of High Performance (ERAD-SE 2025).
The event was hosted by the Institute of Computing at the Universidade Federal Fluminense (UFF) in Niterói, Brazil, from November 5-7, 2025.
The work was authored by Alan L. Nunes (UFF), Lucas A. F. da Costa (CESAR), Felipe A. Portella (Petrobras), José Viterbo (UFF), Leonardo V. Neri (CESAR), Paulo J. B. Estrela (Petrobras), Luciana de C. Franci (CESAR), Renzo Q. Malini (Petrobras), and Lúcia Maria de Assumpção Drummond (UFF).
The paper “Spotting the Right Cloud Instances with Multiple AWS EC2 Fleets” was accepted and successfully presented at the prestigious 37th IEEE/SBC International Symposium on Computer Architecture and High Performance Computing (SBAC-PAD 2025). The conference was held in Bonito, Mato Grosso do Sul, from October 28 to 31, 2025.
The work was authored by students Daniel Sodré (UFF), Miguel de Lima (UFF), and Lucas Serrano (UFF), and professors Cristina Boeres (UFF), Vinod E.F. Rebello (UFF), and Lúcia M. A. Drummond (UFF).
he paper “A Multi-Cloud Approach to Cost Optimization with AWS and Azure Fleet Services” was presented at the Workshop on Cloud Computing (WCC 2025). The event took place in Bonito, Mato Grosso do Sul, between October 28 and 31, 2025, in conjunction with SBAC-PAD.
The work was authored by students Lucas Serrano (UFF) and Miguel Freitas (UFF), Professor Lúcia Drummond (UFF), and Felipe Portella (Petrobras).
Two papers from our research group have been accepted for publication and presentation at the X Southeast Regional School of High Performance (ERAD-SE 2025).
The event will be hosted by the Institute of Computing at the Universidade Federal Fluminense (UFF), in Niterói, Brazil, from November 5 to 7, 2025.
The accepted works are:
The paper was authored by Bernardo Gallo (UFF), Matheus Marotti (UFF), Lúcia Maria de Assumpção Drummond (UFF), José Viterbo (UFF), Felipe A. Portella (PETROBRAS), Paulo J. B. Estrela (PETROBRAS), and Renzo Q. Malini (PETROBRAS).
This work was authored by Alan L. Nunes (UFF), Lucas A. F. da Costa (CESAR), Felipe A. Portella (Petrobras), José Viterbo (UFF), Leonardo V. Neri (CESAR), Paulo J. B. Estrela (Petrobras), Luciana de C. Franci (CESAR), Renzo Q. Malini (Petrobras), and Lúcia Maria de Assumpção Drummond (UFF).
The paper “Towards a Novel Vertical Scaling Approach for Bursty Workloads in Kubernetes” has been accepted for publication and presentation at the 14th International Workshop on Cloud and Edge Computing, and Applications Management (CloudAM 2025).
The work was authored by student Miguel de Lima (UFF), researcher Luan Teylo (Inria), and Professor Lúcia Drummond (UFF). The event will be held in Nantes, France, from December 1st to 4th, 2025.
The paper “A Preliminary Study on Reinforcement Learning and Genetic Algorithm for Job Scheduling” was presented on October 7 at the SBrazilian Symposium on Operational Research (SBPO) 2025. The event took place in Gramado, Rio Grande do Sul, from October 5 to 9.
The work was authored by student Bernardo Gallo (UFF) and professors Lúcia Drummond (UFF) and José Viterbo (UFF).
The paper “Mitigating Concept Drift in Job Execution Time Prediction Models” was presented on September 30 at the Workshop on Knowledge Discovery, Data Mining, and Learning (KDMiLe) 2025. The event took place in Fortaleza, Ceará, from September 29 to October 2.
The work was authored by student Bernardo Gallo (UFF) and professors Lúcia Drummond (UFF) and José Viterbo (UFF).