Program Overview

  1. Home
  2. Program Overview

Program Overview

English EN Portuguese PT
Oct, 28 Oct, 29 Oct, 30 Oct, 31
Kahura Kadiweu Aquidaban Kahura Kadiweu Aquidaban Kahura Kadiweu Aquidaban Kahura Kadiweu Aquidaban
07:30–08:30 Registration Registration Registration Registration
08:30–10:30 SSCAD S1 PT WAMCA S1 EN Tutorial 1: Vinicius Batista (AWS) PT SBAC-PAD S2 EN SSCAD S3 PT WEAC S1 PT SBAC-PAD S4 EN Marathon PT SSCAD S5 PT WIC S4 PT SBAC-PAD S6 EN SSCAD S6 PT Tutorial 4: Ricardo Ferreira (UFV) PT
10:30–10:50 Coffee-break Coffee-break Coffee-break Coffee-break
11:00–12:00 Tutorial 1: Vinicius Batista (AWS) – cont. PT Keynote – Prof. Jesus Carretero EN Keynote – Prof. David Bader EN SSCAD S8 PT SSCAD S7 PT Tutorial 4 – Ricardo Ferreira (UFV) – cont. PT
12:00 Lunch Lunch Lunch Closing Ceremony – Lunch EN
02:00pm SSCAD S2 PT WAMCA S2 EN SSCAD S4 PT Meeting: CE-CC/SBC EN Tutorial 2: Sandro Rigo (UNICAMP) PT SBAC-PAD S5 EN Marathon PT LeanDL – S1 EN Tutorial 3: Aleardo Manacero (UNESP) PT
02:30pm WCC S1 EN
03:00pm Industrial Talk 1 – AWS PT
04:00–04:20pm Coffee-break Coffee-break Coffee-break
04:30pm SBAC-PAD S1 EN WIC S1 PT CTD S1 PT SBAC-PAD S3 EN WCC S2 EN Tutorial 2: Sandro Rigo (UNICAMP) – cont. PT CE-ACPAD Meeting PT Marathon PT LeanDL – S2 EN Tutorial 3: Aleardo Manacero (UNESP) – cont. PT
05:30pm WIC S2 PT CTD S2 PT WIC S3 PT
06:30-07:30pm Keynote – Prof. Manish Parashar EN Industrial Talk 2 – DELL/NVIDIA PT Industrial Talk 3 – Versatus HPC PT
07:30pm Opening Ceremony EN
08:00pm Cocktail PT
08:00pm Conference Dinner EN

Program Details

Oct, 28

Registration

  • Registration

  • Kahura SSCAD S1 – Arquitetura e Sistemas Heterogêneos
    Session chair: Liana Duenha (UFMS)
    • Simulador Web para o Ensino de Arquitetura de Computadores com Suporte a Vetores e Cache
      Gerson Geraldo H. Cavalheiro (UFPEL), Andre Rauber Du Bois (UFPEL), João Linares (UFPEL)
    • Evaluating Memory Constraints of RISC-V Matrix Accelerators using gem5
      Iago Caran Aquino (UNICAMP), Casio Krebs (UNICAMP), Lucas Wanner (UNICAMP), Sandro Rigo (UNICAMP)
    • Integration and performance analysis of parallel RISC-V architectures
      Casio Krebs (UNICAMP), Guido Araujo (UNICAMP), Lucas Wanner (UNICAMP)
    • MSE: A Matrix Sparsity Extension for RISC-V
      Luc Joffily Ribas (UNICAMP), Iago Caran Aquino (UNICAMP), Isaías Felzmann (UNICAMP), Lucas Wanner (UNICAMP), Guido Araujo (UNICAMP)
    • Otimizando Estruturas de Grafos em Memória Persistente para Arquiteturas NUMA
      Lucas Spagnol (UNESP), Alexandro Baldassin (UNESP), Emilio Francesquini (UFABC), Bruno Chinelato Honorio (UNESP), Otavio Scarparo Souza (UNESP)
    • To Pin or not to Pin: That is the question
      Guilherme Galante (UNIOESTE), Marcio Oyamada (UNIOESTE)
  • Kadiweu WAMCA S1
    • Declarative Adaptive Optimization of Task-Based Applications on Heterogeneous Architectures
      Emanuele De Angelis, Guglielmo De Angelis, Romolo Marotta, Federica Montesano, Alessandro Pellegrini and Maurizio Proietti
    • From Static to Quasi-Dynamic: Reconsidering Scheduling and Memory in SDF Compilers
      Pedro Ciambra, Anaëlle Cloarec, Hervé Yviquel, Mickaël Dardaillon and Maxime Pelcat
    • Evaluating Parallelism Strategies and Scheduling for Irregular Problems: A Case Study with OpenMP
      Paulo Zimpel, Vinicius Dias and Samuel Ferraz
  • Aquidaban Tutorial 1: Research Computing with High Performance and Accelerated Computing – Vinicius Batista (AWS)

Coffee-break

  • Coffee-break

  • Aquidaban Tutorial 1: Research Computing with High Performance and Accelerated Computing – Vinicius Batista (AWS) – cont.

Lunch

  • Lunch

  • Kahura SSCAD S2 – Arquitetura, Aplicações e Ferramentas
    • Session chair: Alexandro Baldassin (UNESP)
    • Dependability Analysis of Weather Monitoring Systems Considering Different Redundancy Mechanisms
      Vinícius G. P. Lima (UFRPE), Ermeson Andrade (UFRPE), Danilo R. B. Araújo (UFRPE)
    • Paralelização da Geração de Constraints Lists em Alinhamentos Múltiplos de Sequências Genéticas
      Mario João JR (UERJ), Alexandre Sena (IME, UERJ), Vinod Rebello (UFF)
    • Large-Scale RISC-V Processor Verification Using Automated Design Inspection and a Generic Simulation Method
      Gabriel Gomes (UNICAMP), Julio Nunes Avelar (UNICAMP), Gabriel Oliveira (UNICAMP), Enzo Bertoloti (UNICAMP), Rodolfo de Azevedo (UNICAMP)
    • HPC on a budget: on the CPU’s impact on dense linear algebra computation with GPUs
      Lucas Assis (UFRGS), Lucas Mello Schnorr (UFRGS)
    • Intrusiveness and Scalability of OMPT-Based Tracing Tools for Task-based OpenMP Applications
      Rayan Raddatz (UFRGS), Lucas Mello Schnorr (UFRGS)
    • Workflow para Alinhamento Exato de Sequências em Sistemas de Processamento de Alto Desempenho
      Rafael Terra (LNCC), Kelen Souza (FAETERJ), Hiago Mayk Gomes de Araujo Rocha (LNCC), Carla Osthoff (LNCC), Diego Carvalho (CEFET-RJ), Kary Ocaña (LNCC)
  • Kadiweu WAMCA S2
    • Efficient SIMD and Shared-Memory Parallelization of 3D Acoustic Wave Propagation Simulation
      Chahinèze Ztoti, Claude Tadonki, Roblex Nana Tchakoute and Hervé Chauris
    • ROPH: A Robust, Optimized, and Parallelized Harris Detector with Flexible FAST-Based Pruning
      Andres Giraldo Morales, Cristiana Bentes, Maria Clicia Castro, Claude Tadonki and Gilson Costa
    • Impact of Data Distribution and Schedulers for the LU Factorization on Multi-Core Clusters
      Otho José Sirtoli Marcondes, Philippe O. A. Navaux and Lucas Schnorr

  • AquidabanIndustrial Talk 1: AWS
    • Title:High Performance Computing na AWS
    • Speaker: Vinicius Batista – AWS

Coffee-break

  • Coffee-break

  • Kahura SBAC-PAD S1 – Computer Architecture
    Session chair: Nelson Amaral (University of Alberta, Canada)
    • TRAP: Time-Aware Probabilistic In-DRAM RowHammer Solution
      Samiksha Verma (Indian institute of Technology Bombay, India), Virendra Singh (Indian institute of Technology Bombay, India)
    • Extraction and Representation of Sparsity Patterns for Efficient Data Transfer on Accelerators
      Yang Su (University of Toronto, Canada), Toshiyuki Ichiba (Fujitsu Limited, Japan), Katsuhiro Yoda, Yasuhiro Watanabe (Fujitsu Limited, Japan), Takahide Yoshikawa, Tarek Abdelrahman (University of Toronto, Canada)
    • MIDAS: A Mapping Infrastructure for Configurable, Data-Streaming based Domain Specific Accelerators
      Martim Bento (Instituto Superior Tecnico / INESC-ID, Portugal), Nuno Neves, Pedro Tomás (INESC-ID, Portugal), Nuno Roma (INESC-ID, Portugal)
    • Heuristics for Energy-Efficient Instruction-Level Approximate Computing
      Liana D. Duenha (Universidade Federal de Mato Grosso do Sul, Brazil), Felipe de Sousa Sovernigo (UFMS, Brazil), Gregório Koslinski Neto (UFMS, Brazil), Daniela Luiza Catelan (UFMS, Brazil), Ricardo Ribeiro dos Santos (UFMS, Brazil)
  • Kadiweu WIC S1
    • Performance Evaluation of Convolutional Neural Networks with oneAPI and OpenMP for Network Intrusion Detection
      Rafael Oliveira (PUC MINAS); Thiago Leão (PUC MINAS); Henrique Cota de Freitas (PUC MINAS)
    • A Performance Analysis of System Rollback Techniques – Antônio Sousa (PUC MINAS); Henrique Cota de Freitas (PUC MINAS)
    • Performance Evaluation of k-means using CPU and GPU with oneAPI and OpenMP for Network Intrusion Detection
      Laura Caetano Costa (PUC MINAS); Luiz Fernando Antunes da Silva Frassi (PUC MINAS); Henrique Cota de Freitas (PUC MINAS)
  • Aquidaban CTD S1
    • Análise de desempenho, custo energético e acurácia de um módulo de um modelo numérico de Previsão Meteorológica usando Precisão Reduzida
      Marcelo Augusto Sudo, Alvaro Luiz Fazenda (Universidade Federal de São Paulo)
    • Improving the energy efficiency of data stream classifier arrays for edge computing
      Reginaldo Luna, Hermes Senger (Universidade Federal de São Carlos)
    • Stealth Metrics: A Less Intrusive I/O Tracer for Application I/O Characterization
      Rodrigo Nascimento, Alfredo Goldman (Universidade de São Paulo)

  • Kadiweu WIC S2
    • Dynamic Load Balancing and Scalability Analysis of the Mandelbrot Set in a Multi-Threaded HPC Application
      Francisco Etcheverria (UFRGS); Rayan Raddatz (UFRGS); Kenichi Brumati (UFRGS); Lucas Mello Schnorr (UFRGS)
    • Uma proposta de monitoramento hierárquico e em anel utilizando heartbeat para a biblioteca DeLIA
      Cleverson Silva (MACKENZIE); Gustavo Santos (MACKENZIE); João Mota (MACKENZIE); Calebe de Paula Bianchini (MACKENZIE)
    • Predição de Custo de Execução de FaaS em Provedor Público de Nuvem por meio do Framework Orama
      Edson Sales (UNB); Leonardo Rebouças de Carvalho (UNB); Aleteia de Araujo (UNB)
  • Aquidaban CTD S2
    • An Intelligent context-aware edge-based data reduction framework for IoT
      Laercio Pioli Junior, Mario Antonio Ribeiro Dantas, Douglas Dyllon Jeronimo de Macedo (Universidade Federal de Santa Catarina)
    • Uma Arquitetura Autoadaptável para a Implantação de Observabilidade em Fog Computing
      Breno Gustavo Soares da Costa, Aletéia de Araujo (Universidade de Brasília)
    • Uma Proposta para a Descoberta e Alocação de Recursos Computacionais em Fog Computing
      Joao Bachiega, Aletéia de Araujo (Universidade de Brasília)

Keynote

  • Keynote: Unleashing Responsible AI for Science: Taming Open Data
  • Speaker: Prof. Manish Parashar – Utah State University
  • Abstract: Artificial intelligence (AI) and open data have become essential engines for scientific discovery and innovation. However, realizing this transformative potential requires a transdisciplinary approach that ensures research and development can effectively and responsibly leverage the diversity of data sources. Despite the exponential growth of available digital data sources and the ubiquity of non-trivial computational power for processing this data, realizing data-driven, AI-enabled science workflows remains challenging. In this talk, I will discuss the importance of democratizing AI R&D, including access to open data and advanced cyberinfrastructure. I will introduce the University of Utah’s One-U Responsible AI Initiative, which aims to catalyze an innovation ecosystem at the University of Utah and across the state. I will also present the vision, architecture, and deployment of the National Data Platform project, as part of a broader national cyberinfrastructure, aimed at catalyzing an open and extensible data ecosystem for science.

  • Opening Ceremony

  • Cocktail

Oct, 29

Registration

  • Registration

  • Kahura SBAC-PAD S2 – Resource Management
    Session chair: Hans-Ulrich Heiss (TU-Berlin, Germany)
    • Data Management in the Continuum: Cross-facility Object-based Data Transfers
      Jean Luca Bez (Lawrence Berkeley National Laboratory, United States of America), Houjun Tang (Lawrence Berkeley National Laboratory, United States of America), Chen Wang (Nanyang Technological University, Singapore), Suren Byna (The Ohio State University, United States of America)
    • Mobility-aware placement of service-composed applications on Cloud-Edge Continuum
      Paulo Roberto Albuquerque, Guilherme Piêgas. Koslovski (UDESC, Brazil), Maurício Pillon (UDESC, Brazil), Tiago Coelho Ferreto (PUCRS, Brazil)
    • Spotting the Right Cloud Instances with Multiple AWS EC2 Fleets
      Daniel Marcondes Bougleux Sodré (UFF, Brazil), Miguel Junior (UFF, Brazil), Lucas Serrano (UFF, Brazil), Cristina Boeres (Instituto de Computacao, UFF, Brazil), Vinod Rebello (UFF, Brazil), Lucia M. A. Drummond (UFF, Brazil)
    • Hierarchical Dynamic Multilevel Graph Partitioning for Load Balancing in Distributed Agent-Based Simulations
      Cristina Peralta Quesada (Universitat Autònoma de Barcelona, Spain), Eduardo Cesar (Universitat Autònoma de Barcelona, Spain), Andreu Moreno Vendrell (EUSS, Spain), Anna Sikora (Universitat Autònoma de Barcelona, Spain)
  • Kadiweu SSCAD S3 – Computação Distribuída e em Nuvem
    Session chair: Márcio Castro (UFSC)
    • Assessing the Performance and Impact of an Open-Source RTI in Distributed Real-Time Military Simulation
      Joao Rodolfo Oliveira Rosa (UNIFESP), Alvaro Luiz Fazenda (UNIFESP)
    • Implementação de Tolerância a Falhas no Método Lattice Boltzmann para Execução Resiliente em Instâncias Efêmeras da AWS
      Rafael Vargas (UFSC), Vanderlei Pereira Filho (UFSC), Márcio Castro (UFSC)
    • Implementing Cold-Start Reduction Techniques on Globus Compute
      João Gabriel Lembo (USP), Alfredo Goldman (USP)
    • A Literature Review on Live Migration in Container-Based Distributed Systems
      João Barroso (UNESP), Aleardo Manacero Jr. (UNESP), Renata Spolon Lobato (UNESP), Roberta Spolon (UNESP)
    • A Weighted Bi-objective Strategy for Executing Scientific Workflows in Containerized Environments
      Wesley Ferreira (UFF), Liliane Kunstmann (IMPA), Yuri Abitbol Frota (UFF), Luan Teylo (INRIA Bordeaux University), Daniel de Oliveira (UFF)
    • Federated Outlier Detection for Astronomical Data: Performance Analysis on Commercial Clouds
      Camila Lopes (UFF), Wesley Ferreira (UFF), Julia Gschwend, Luiz Nicolaci da Costa (LIneA), Rafael Ferreira da Silva (ORNL), Marta Mattoso (COPPE/UFRJ), Aline Paes (UFF), Daniel de Oliveira (UFF)
  • Aquidaban WEAC S1
    • Teaching Computer Architecture through an Integrated Top-Down RISC-V Processor Design Approach
      Guilherme Esmeraldo (IFCE), Edson Lisboa (IFS), Victor Medeiros (UFPE), Edna Barros (UFPE)
    • FPGA Unboxing: entendendo a arquitetura e as ferramentas de projeto para FPGAs
      Deborah Caroline Rodrigues Oliveira (UFOP), João Silva (UFOP), José Augusto Miranda Nacif (UFV), Ricardo dos Santos Ferreira (UFV), Racyus Delano Garcia Pacífico (UFOP)
    • Retrieval-Augmented Large Language Models for Computer Architecture Learning and Design Assistance
      Wenderson Júnio de Souza (PUC MINAS), Humberto Torres Marques Neto (PUC MINAS), Henrique Cota Freitas (PUC MINAS)
    • VeryGA
      Interface Modular VGA para Simulação de Verilog       – Talles de Sousa Costa (UFV), Racyus Delano Garcia Pacífico (UFOP), Ricardo dos Santos Ferreira (UFV)
    • Desenvolvendo simuladores para arquitetura de computadores com auxílio de modelos generativos de linguagens
      Racyus Delano Garcia Pacífico (UFOP), Ricardo dos Santos Ferreira (UFV)

Coffee-break

  • Coffee-break

Keynote

  • Keynote: Dynamic Resource Management for Next-Gen HPC Applications and Architectures
  • Speaker: Prof. Jesus Carretero – UC3M
  • Abstract: The current static usage model of HPC systems is becoming increasingly inefficient. This is driven by the continuously growing complexity and heterogeneity of system architectures, in combination with the increased usage of coupled applications, the need for strong scaling with extreme scale parallelism, and the increasing reliance on complex and dynamic workflows. As a consequence, we see a rise in research on malleable systems, middleware software and applications, which can adjust resources usage dynamically in order to extract a maximum of efficiency. By providing an intelligent global coordination of resources usage, through runtime scheduling of computation, network usage and I/O across all components of the system architecture, malleable HPC systems can maximize the exploitation of their resources, while at the same time minimizing the makespan of applications in many, if not most, cases. Such malleable systems, however, face a series of fundamental research challenges, including: who initiates changes in resource availability or usage? How is it communicated? How to compute the optimal usage? How can applications cope with dynamically changing resources? What should malleable programming models and abstractions look like? How to design resource management frameworks for malleable systems? Which resources benefit from malleability and which (if any) should still be managed statically? This Keynote will present the state of the art in dynamic resource management for HPC systems and will address the former question and some possible solutions to them.

Lunch

  • Lunch

  • Kahura SSCAD S4 – Aprendizado de Máquina
    Session chair: Vinicius Dias (UFLA)
    • Predicting FaaS Runtime with the Orama Framework Using Machine Learning
      Leonardo Rebouças de Carvalho (UnB), Geraldo Rocha (UESB), Aleteia de Araujo (UnB)
    • Otimização de Parâmetros de Buffer Pool com Aprendizado de Máquina em Ambientes Não Transacionais
      Eduardo Pingarilho Mendizabal (UnB), Geraldo Rocha (UESB), Aleteia de Araujo (UnB)
    • Evaluating Machine Learning Algorithms for Anomaly Detection in Industrial Engines on Edge Devices
      Lucas Edson Silva de Araújo (UFRPE), Sergio Chevtchenko (UFRPE), Danilo R. B. Araújo (UFRPE), Ermeson Andrade (UFRPE)
    • Quantum Machine Learning with Enhanced Autoencoders for Intrusion Detection
      Milleny Teixeira de Souza (PUC Minas), Matheus Alcântara Souza (PUCMG), Henrique Cota de Freitas (PUC Minas)
    • Adaptive Detection of Software Aging under Workload Shift
      Rafael José Moura da Silva (UFRPE), Maria Gizele Alves do Nascimento (UFRPE), Fumio Machida, Ermeson Andrade (UFRPE)
    • Analisando Técnicas de Gestão de Energia em Aplicações Aceleradas por GPU em um Sistema Exascale
      Mariana Costa (PPGC/UFRGS), Antonio Tadeu Azevedo Gomes (LNCC), Philippe Olivier Alexandre Navaux (UFRGS), Bronson Messer, Arthur Francisco Lorenzon (UFRGS)
  • Kadiweu Meeting: Meeting: SBC’s Special Comission on Cloud Computing (CE-CC/SBC)
  • Aquidaban Tutorial 2: Implementing a new RISC-V Instruction with LLVM Compiler Infrastructure – Sandro Rigo (UNICAMP)

  • Kadiweu WCC Session 1
    • Opening session – WCC 2025
    • A Multi-Cloud Approach to Cost Optimization with AWS and Azure Fleet Services
      Lucas Serrano, Miguel de Lima, Lúcia Drummond and Felipe Portella
    • Can GPUs help scaling traditional Apache Spark workloads?
      Moises Felipe Lehnen, Lucas Mello Schnorr and Philippe Olivier Alexandre Navaux
    • Revisiting Gradient Staleness: Evaluating Distance Metrics for Asynchronous Federated Learning Aggregation
      Patrick Wilhelm and Odej Kao

Coffee-break

  • Coffee-break

  • Kahura SBAC-PAD S3 – Parallel Applications and Algorithms
    Session chair: Claude Tadonki (Mines ParisTech, France)
    • Evaluating Code Portability for Carbon-Efficient RTM Computing
      Arthur Francisco Lorenzon (UFRGS, Brazil), Alexandre Sardinha (Petróleo Brasileiro S.A., Brazil), Philippe Olivier Alexandre Navaux (UFRGS, Brazil), Bronson Messer (Oak Ridge National Laboratory, USA)
    • A Distributed and Storage-Aware Approach to Large-Scale Cholesky Factorization
      Carla Cardoso (UNICAMP, Brazil), Rodrigo Ceccato de Freitas (UNICAMP, Brazil), Sandro Rigo (UNICAMP, Brazil), Guido Araujo (UNICAMP, Brazil), Herve Yviquel (UNICAMP, Brazil)
    • Superstencil: A Memory-Efficient Superstep Wave Propagation Method for Seismic Imaging
      George Gigilas Junior (Universidad Estadual de Campinas, Brazil), Pedro da Silva Peixoto (USP, Brazil), Hermes Senger (UFSCAR, Brazil), Hervé Yviquel (UNICAMP, Brazil)
    • DynaMap: A Map Equation-based Parallel Algorithm for Detecting Communities on Dynamic Graphs
      Gabriel Giordani dos Santos (PUCRS, Brazil), Kartik Lakhotia, Cesar De Rose (PUCRS, Brazil)
  • Kadiweu WCC S2
    • Performance and Cost Analysis of AWS Burstable Instances for HPC with NAS Parallel Benchmarks
      Artur Luiz Rizzato Toru Soda, Vanderlei Munhoz Pereira Filho and Márcio Castro
    • Leveraging Large Language Models for Anomaly Detection in Microservices Architectures
      Diego Pedroso, Luís Almeida, William Akihiro Aisawa, Inês Dutra and Sarita Mazzini Bruschi
    • Practical Anomaly Detection for Infrastructure KPIs under Real-Time Constraints
      Maynara Natalia Scoparo and Sarita Mazzini Bruschi
    • Closing session
  • Aquidaban Tutorial 2: Implementing a new RISC-V Instruction with LLVM Compiler Infrastructure – Sandro Rigo (UNICAMP) – cont.

  • Kadiweu WIC S3
    • Análise Comparativa do Algoritmo de Shor em Arquiteturas Clássica e Quântica
      Caroline Gandolfi (IME-RJ); Evelyn Henriques Q. V. Costa (IME-RJ); Anderson Fernandes Pereira dos Santos (IME-RJ); Raquel Coelho Gomes Pinto (IME-RJ); Vitor S. M. Sakai (IME-RJ)
    • Avaliação de Algoritmos de Ordenação de Dados em Ambiente de HPC com Raspberry Pi
      Lucayan Felipe Teixeira da Silva (IFRO); Wanderson Roger Azevedo Dias (IFRO)
    • Comparação de Paralelismo com DO CONCURRENT, OpenMP e MPI em Algoritmos do NAS Parallel Benchmark
      Anna Victória Gonçalves Marciano (UNIPAMPA); Artur dos Santos Antunes (UNIPAMPA); Claudio Schepke (UNIPAMPA)

– Industrial Talk: DELL/NVIDIA

  • Title: AMD EPYC 5ª Gen for Scientific/AI (ML) HPC
  • Speaker: Alexandre Ventura de Moraes (DELL/NVIDIA)
  • Abstract: As soluções Dell PowerEdge de 17ª geração com processadores AMD EPYC™ 9000 (5ª Gen) e aceleradores AMD Instinct™ MI300X para HPC/IA/ML. Alta densidade de núcleos (até 128), Boost On, suporte a FMA32 (duas AVX-512 por ciclo) e até 1,5 TB de HBM3 por servidor no XE9680 com MI300X. Ecossistema aberto, escalável e gerenciável via Dell Omnia e AMD ROCm™.

Oct, 30

Registration

  • Registration

  • Kahura SBAC-PAD S4 – HPC for AI
    Session chair: Luiz Fernando Bittencourt (Unicamp)
    • Generative Fabrication of Medical Images for Machine Learning Training
      Andres G. Calzada-Jasso, Andrei Tchernykh (CICESE RESEARCH CENTER, Mexico), Ixchel D. Avendaño-Pacheco, Jorge M. Cortés-Mendoza (National College of Ireland, Ireland), Bernardo Pulido-Gaytan (National College of Ireland, Ireland), Mikhail Babenko, Alfredo Goldman (USP, Brazil), Horacio Gonzalez-Velez (National College of Ireland, Ireland)
    • Scalable and Efficient Deep Learning for Diabetic Retinopathy Classification on ARM
      Thiago Da Silva Araújo (UFRGS, Brazil), Philippe Olivier Alexandre Navaux (UFRGS, Brazil), Beatriz Schaan (UFRGS, Brazil), Carla Maria Dal Sasso Freitas (UFRGS, Brazil)
    • SPINN: a Tool for Distributed Patch Inference on Massive Data Samples
      João Seródio (UNICAMP, Brazil), Julio Cesar Faracco (UNICAMP, Brazil), Fernando Gubitoso Marques (UNICAMP, Brazil), Otavio Napoli (University of Campinas, Brazil), Alan Souza, Daniel Miranda, Carlos Alberto Astudillo Trujillo (UNICAMP, Brazil), Edson Borin (University of Campinas, Brazil)
    • Accelerating GNN Inference via Automated Parallel Execution on Edge Heterogeneous Platforms
      Yi-Chien Lin (University of Southern California, United States of America), Haoyang Fan (University of Southern California, United States of America), Sameh Gobriel (Intel Labs, United States of America), Nilesh Jain (Intel Labs, United States of America), Viktor Prasanna (University of Southern California, United States of America)
  • Kadiweu Marathon
  • Kadiweu SSCAD S5 – Desempenho, Otimização e Escalabilidade
    Session chair: Arthur Lorenzon (UFRGS)
    • Performance Evaluation of N-Body Simulations on AWS with StarPU, OpenMP and MPI Runtime Systems
      Nicolas Vanz (UFSC), Vanderlei Pereira Filho (UFSC), Márcio Castro (UFSC)
    • NUMA-Aware Task Scheduling Strategy Aiming to Reduce Cache Conflicts
      Thiago de Campos Ribeiro Nolasco (PUC Minas), Pedro Henrique Penna (Microsoft Research), Henrique Cota de Freitas (PUC Minas)
    • Can OpenMP Scale Beyond the Node? A Performance Evaluation of Remote Offloading via the MPI Proxy Plugin
      Jhonatan Cléto (UNICAMP), Guilherme Valarini (CTI Renato Archer), Hervé Yviquel (UNICAMP)
    • Implementação e Avaliação de Políticas de Escalonamento de Warps no Vortex uma GPGPU de Código Aberto
      Samuel Augusto Oliveira Magalhães (CEFETMG), Poliana A. C. Oliveira (CEFETMG), Renan Albuquerque Marks (UFMS)
    • Fast-Tracking Scalability Analysis: The PaScal Paramount Approach
      Reilta Christine Dantas Maia (UFRN), Samuel Xavier de Souza (UFRN)
    • A Client-Side Architecture for Enhancing Productivity of Interactive Parallel Scalability Analysis with PaScal Suite
      Igor Sérgio de França Correia (UFRN), Samuel Xavier de Souza (UFRN)
  • Aquidaban WIC S4
    • HeatSync: Sistema de Refrigeração Externa Automatizado para Notebooks
      Luís Augusto Lima de Oliveira (PUC MINAS); Gabriel Mourão (PUC MINAS); Rafael Oliveira (PUC MINAS); Victor Ferraz de Moraes (PUC MINAS); Mateus Barbosa (PUC MINAS); Rafael Henriques Nogueira Diniz (PUC MINAS); Matheus Pereira (PUC MINAS); Felipe Domingos da Cunha (PUC MINAS); Matheus Alcântara Souza (PUC MINAS)
    • Avaliação do Consumo Energético e Desempenho de Clusters Baseados em Banana Pi e Raspberry Pi
      Luís Fillipe Pereira (PUC MINAS); Henrique Cota de Freitas (PUC MINAS)
    • Desenvolvimento de Simulador Web para Ensino de Arquitetura: O Caso Neander-V
      João Linares (UFPel); Andre Rauber Du Bois (UFPel); Gerson Geraldo H. Cavalheiro (UFPel)
    • Processor CI Inspector: Detectando parâmetros automaticamente em processadores RISC-V
      Gabriel Oliveira (UNICAMP); Julio Nunes Avelar (UNICAMP); Enzo Bertoloti (UNICAMP); Gabriel Gomes (UNICAMP); Rodolfo de Azevedo (UNICAMP)
    • Shell interativo com carregador para RISC-V em FPGA
      Luiz Vartuli (UNB); Marcus Vinicius Lamar (UNB); Alba Cristina Magalhaes Melo (UNB)
    • Gerenciamento de Conflito SPI no Kit BitDogLab: Uma Abordagem com Estados Persistentes em RAM e Soft Resets
      Danielly Almeida (IFMA); Priscila Lima Rocha (IFMA); Marcony Henrique Bento Souza (IFMA)
    • Análise e Comparação de Estratégias de Implementação Física de Multiplicadores Matriciais
      Rafael Ramos de Aguiar (UNICAMP); Isaías Felzmann (UNICAMP)

Coffee-break

  • Coffee Break

Keynote

  • Keynote: High-Performance Graph Analytics for Motif Finding in Neuroscience Connectome Graphs and Beyond using Arachne
  • Speaker: Prof. David Bader – NJIT
  • Abstract: The growth of network-structured data across domains like neuroscience and cybersecurity demands scalable graph analytics, but complex tasks like subgraph isomorphism remain accessible only to high-performance computing (HPC) specialists. Arachne is an open-source framework that democratizes high-performance graph analytics through a Python interface while abstracting parallelism complexities. It enables advanced graph algorithms to run efficiently from laptops to supercomputers. Arachne has been adopted by Harvard researchers for the MoMo connectome visualization tool, allowing neuroscientists to draw neural motifs that are translated into attributed subgraphs and searched using our novel HiPerMotif algorithm. Key innovations include HiPerMotif, which achieves up to 66× speedups over parallel approaches.Testing on large-scale datasets including FlyWire and the H01 human brain connectome demonstrates Arachne’s performance: completing complex subgraph searches in 38 seconds versus NetworkX’s 16,000+ seconds. This unified platform balances high-performance computation with accessibility, enabling researchers to extract insights from billion-scale graphs and advancing pattern matching across data-driven sciences. This research is supported in part by NSF grants CCF-2109988, OAC-2402560, and CCF-2453324.

Lunch

  • Almoço

  • Kahura SBAC-PAD S5 – Performance Evaluation
    Session chair: Tiago Ferreto (PUCRS)
    • Towards Portability at Scale: A Cross-Architecture Performance Evaluation of a GPU-enabled Shallow Water Solver
      Johansell Villalobos (Centro Nacional de Alta Tecnología, Costa Rica), Esteban Meneses, Silvio Rizzi (Argonne National Laboratory, United States of America), Daniel Caviedes-Voullième (FZJ, Germany)
    • Fine-grained Communication Phase based Analytical Performance Modeling and Analysis
      Vishal Deka (Indian Institute of Technology, Kanpur, India), Preeti Malakar (IIT Kanpur, India)
    • Performance, Portability, and Productivity of HIP on GPUs with NAS Parallel Benchmarks
      Gabriell Alves de Araujo (PUCRS, Brazil), Dalvan Griebler (PUCRS, Brazil), Luiz Gustavo Leão Fernandes (PUCRS, Brazil)
    • A Framework for Analytical Performance and Energy Prediction of DL Training on GPUs
      Roblex Nana Tchakoute (Ecole des Mines de Paris, France), Claude Tadonki (MPPSL, France), Petr Dokladal (Ecole des Mines de Paris, France), Youssef Mesri
  • Kadiweu Marathon
  • Kadiweu LeanDL – S1
    • Experimental Evaluation of Quantization Methods in Facial Recognition
      Carlos Monteiro (UFMS, Brazil), Evandro Raphaloski (Smar Equipamentos Industriais Ltda, Brazil), Edson Takashi Matsubara (Fundação Universidade Federal de Mato Grosso do Sul, Brazil)
    • Energy-Aware Deep Learning on GPUs through Parameter Sharing and Mixed Precision Training
      Roblex NANA TCHAKOUTE (Ecole des Mines de Paris, France), Claude TADONKI (MPPSL, France)
    • Text2Graph: Combining Lightweight LLMs and GNNs for Efficient Text Classification in Label-Scarce Scenarios
      Joao Luz (USP/ICMC, Brazil), Ricardo Marcondes Marcacini (ICMC/USP, Brazil)
  • Aquidaban Tutorial 3: Tuning e depuração de aplicações em OpenMPI 5.0 – Aleardo Manacero (UNESP)

Coffee-break

  • Coffee-break

  • Kahura CE-ACPAD Meeting
  • Kadiweu Marathon
  • Aquidaban LeanDL – S2
    • Experimental Reducing Costs in Large-Scale Classification: A Hybrid BERT–LLM Strategy
      Augusto Miranda (UFMS) , Matheus Yasuo Ribeiro Utino (Universidade de São Paulo/ICMC, Brazil), Marcos Gôlo (University of São Paulo, Brazil), Marcela Santos , Mariana Caravanti de Souza (UFMS, Brazil)
    • Applying Large-Scale Model Adaptation Techniques
      Kenzo Sakiyama (USP, Brazil), Magaly L. Fujimoto (USP, Brazil), René Santin (USP, Brazil), Solange Rezende (USP/ICMC, Brazil) (MPPSL, France)
    • One-Class Lightweight Interpretable Filtering For Academic Profiles and Strategic Themes Affinity
      Marcos Gôlo (University of São Paulo, Brazil), Matheus Yasuo Ribeiro Utino (Universidade de São Paulo/ICMC, Brazil) (ICMC/USP, Brazil)
    • Efficient Filtering with BERT Embeddings for Researcher: Topic Affinity Prediction in HPC Pipelines
      Matheus Yasuo Ribeiro Utino (Universidade de São Paulo/ICMC, Brazil), Marcos Gôlo (University of São Paulo, Brazil)
  • Aquidaban Tutorial 3: Tuning e depuração de aplicações em OpenMPI 5.0 – Aleardo Manacero (UNESP) – cont.

– Industrial Talk: Versatus HPC

    TBD
  • Speaker: Eiji Kawahira (Versatus)
  • Abstract: TBD

  • Conference Dinner

Oct, 31

Registration

  • Registration

  • Kahura SBAC-PAD S6 – System Software
    Session chair: Lucia Drummond (Federal Fluminense University, Rio de Janeiro, Brazil)
    • Obstruction-Free Software Transactional Memory for GPUs
      Tiago Perlin (UFPEL, Brazil), Andre Rauber Du Bois (UFPEL, Brazil), Gerson Geraldo H. Cavalheiro (UFPEL, Brazil)
    • A-Flow: managing dataflows on the computing continuum using abstract communication channels
      Catherine Torres Charles (Universidad Carlos III de Madrid, Spain), Dante Domizzi Sanchez Gallegos (Universidad Carlos III de Madrid, Spain), Diana Carrizales-Espinoza, Jose L. Gonzalez-Compean, Jesus Carretero (Universidad Carlos III de Madrid, Spain)
    • Efficient Multi-Workload Execution for Sustainable GPU Performance
      Matheus Costa (UFRGS, Brazil), Philippe Olivier Alexandre Navaux (UFRGS, Brazil), Silvio Rizzi (ANL, United States of America), Bronson Messer, Arthur Francisco Lorenzon (UFRGS, Brazil)
    • Profiler-Guided Execution of Recurrent OpenMP Task Graphs on Heterogeneous Clusters
      Rémy Neveu (Institute of Computing, UNICAMP, Brazil), Rodrigo Ceccato de Freitas (UNICAMP, Brazil), Adrián Munera (Barcelona Supercomputing Center, Spain), Sara Royuela, Jose M. Monsalve Diaz, Hervé Yviquel (UNICAMP, Brazil)
  • Kadiweu SSCAD S6 – Computação Sustentável e Eficiência Energética
    Session chair: Álvaro Fazenda (Unifesp)
    • Estimating CO2 emissions of distributed applications and platforms with SimGrid/Batsim
      Gabriella Saraiva (USP), Miguel Felipe Silva Vasconcelos (IRIT), Sarita Mazzini Bruschi (USP), Danilo Carastan dos Santos (UGA), Daniel Cordeiro (USP)
    • Compressão de Dados para Redução de E/S em Simulações Sísmicas de Alto Desempenho
      Cristiano Alex Künas (UFRGS), Gabriel Freytag (UFSM), Thiago Da Silva Araújo (UFRGS), Rodrigo Machado (PPGC/UFRGS), Bruno Machado Morales (UFRGS), Alexandre Sardinha (Petrobras), Philippe Olivier Alexandre Navaux (UFRGS)
    • Otimização de Fábricas de IA Sustentáveis por Compartilhamento de GPU
      Matheus Costa (UFRGS), Sandro Rigo (UNICAMP), Carla Osthoff (LNCC), Silvio Rizzi (ANL), Philippe Olivier Alexandre Navaux (UFRGS), Arthur Francisco Lorenzon (UFRGS)
    • Extração Eficiente de MFCCs em FPGA: Uma implementação Aberta e Flexível
      Julio Nunes Avelar (UNICAMP), Vinicius Patriarca Miranda Miguel (UNICAMP), Tiago Zaparoli (UNICAMP), Enzo Bertoloti (UNICAMP), Gabriel Oliveira (UNICAMP), Rodolfo de Azevedo (UNICAMP)
    • Modelagem Preditiva de EDP para Otimização de Submissões em Supercomputadores
      Micaella C. V. Paula (LNCC), Alexandre L. Porto (LNCC), Hiago Mayk Gomes de Araujo Rocha (LNCC), Isabella Muniz (LNCC), Douglas Cardoso (CEFET-RJ), Kary A. C. S. Ocaña (LNCC), Arthur Francisco Lorenzon (UFRGS), Philippe Olivier Alexandre Navaux (UFRGS), Carla Osthoff (LNCC)
    • PowerTrackZ: Análise da eficiência energética de pontos de acesso em redes sem fio
      Pedro Acácio Rodrigues (UTFPR Campo Mourão), João Fabrício Filho (UTFPR Campo Mourão)
  • Aquidaban Tutorial 4: Mais Produtividade com LLMs, Engenharia de Prompt, Aprendizado de Máquina e GPUs – Ricardo dos Santos Ferreira (UFV)

Coffee-break

  • Coffee-break

  • Kahura SSCAD S8 – Algoritmos Paralelos e Distribuídos I
    Session chair: Lucas Schnorr (UFRGS)
    • Interface para Programação de Pipelines Lineares Tolerantes a Falha para MPI Padrão C++
      Eduardo Martins (PUCRS), Renato Barreto Hoffmann Filho (PUCRS), Lucas Alf (PUCRS), Dalvan Griebler (PUCRS)
    • Preventing Out-Of-Memory Errors in Dask through Automated Memory-Aware Chunking
      Daniel Fonseca (UNICAMP), Edson Borin (UNICAMP), Carlos Alberto Astudillo Trujillo (UNICAMP)
    • Superando Limites no Multiparticionamento em GPU
      Michel B. Cordeiro (UFPR), Wagner M. Nunan Zola (UFPR)
  • Kadiweu SSCAD S7 – Algoritmos Paralelos e Distribuídos
    Session chair: Luiz Bittencourt (Unicamp)
    • Investigando Gerenciamento de Contenção em um Sistema de Memória Transacional Distribuída
      Rafael Rutz dos Santos (UFPEL), Gerson Geraldo H. Cavalheiro (UFPEL), Andre Rauber Du Bois (UFPEL)
    • An Experimental Study of Variable Neighborhood Search for General-Purpose Subgraph Optimization in Parallel Systems
      Diogo Oliveira Carvalho (UFLA), Mayron César de Oliveira Moreira (UFLA), Vinícius Dias (UFLA)
    • Revisitando Clássicos da Concorrência: Implementação e Avaliação em OpenMP, Rust e Go
      Lucas Braatz Araujo (UFPEL), Daniel Di Domenico (UFPEL), Andre Rauber Du Bois (UFPEL), Gerson Geraldo H. Cavalheiro (UFPEL)
  • Aquidaban Tutorial 4: Mais Produtividade com LLMs, Engenharia de Prompt, Aprendizado de Máquina e GPUs – Ricardo dos Santos Ferreira (UFV) – cont.

Closing Ceremony – Lunch

  • Closing Ceremony – Lunch

Sponsors:

Diamond:

Gold:

Silver:

Partner:

Organized by:

Promoted by:

Funded by: