Home / Program
Wednesday 02 October
08:30 - 18:00
Registration
Please remember to present the QR code you received via email.
Main Courtyard, d’Etigny Auditorium
09:00 - 10:00
Sponsors: Lenovo+AMD+NVIDIA
09:00 - 09:45
Welcome
Main Courtyard, d’Etigny Auditorium
09:45 - 10:45
Keynote Talk – Bernd Mohr
Parallel Performance Analysis at Scale: From Single Node to one Million HPC Cores
Current high-end HPC systems consist of complex configurations of potentially heterogeneous componen ...
Main Courtyard, d’Etigny Auditorium
10:00 - 11:00
GitOps en HPC
Nicolas Alejandro Kowenski, (ETHZ - Eidgenössische Technische Hochschule Zürich - Suiza)
10:45 - 11:05
Industrial Talk Diamond – Dell+AMD+NVIDIA
Main Courtyard, d’Etigny Auditorium
11:00 - 11:20
Sponsor: ThinLinc
11:05 - 11:45
Coffee Break and Industrial Expo
Enrique d'Etigny Auditorium Break Room - Below Auditorium
11:20 - 12:00
Coffee
11:45 - 12:25
Invited Talk – Alberto Ros
Timely Prefetching for High Performance Computers
Prefetching instructions and data is a fundamental technique for designing high-performance computer ...
Main Courtyard, d’Etigny Auditorium
12:00 - 13:00
Maximizando la Productividad en HPC: Diseño de Políticas SLURM en CINECA
Orlenys Natali Troconis (CINECA - Consorzio Interuniversitario per il Calcolo Automatico – Italia)
12:25 - 12:45
Industrial Talk Platinum – Intel
Main Courtyard, d’Etigny Auditorium
12:45 - 14:15
Lunch Time
Av. Almte. Blanco Encalada 2540, Santiago, Región Metropolitana
Club House, Club Hípico, Santiago.
13:00 - 14:30
Almuerzo
14:15 - 15:15
Special Talk – Mateo Valero
Latest Trends in High Performance Computer Architectures
In 1956, the term Artificial Intelligence was coined. From then on, research began that was thought ...
14:30 - 15:15
Vigilar y Castigar: el rol del monitoreo en HPC
Orlenys Natali Troconis (CINECA - Consorzio Interuniversitario per il Calcolo Automatico – Italia)
15:15 - 16:00
Hardening en nodos de login y cómputo en Rocky Linux 9, Sub y Sobreutilización de recursos en HPC
Pablo Flores Aravena (NLHPC - Laboratorio Nacional de Computación de Alto Rendimiento - Chile)
15:15 - 16:15
Panel – Government and Public Relations
The main objective of the panel is to understand the strategies and plans of different governments i ...
Main Courtyard, d’Etigny Auditorium
16:00 - 16:30
Coffee
16:15 - 16:45
Poster Session – Opening
16:15 - 16:45
Coffee break, Posters and Expo
16:45 - 18:45
HPC Track
16:30-16:50 - A Comprehensive Analysis of Process Energy Consumption on Multi-Socket Systems with GPUs
16:50-17:10 - Enhancing Reverse Time Migration Simulations in HPC Systems through I/O and Computation Overlapping
17:10-17:30 - MKL Vs cuBLAS
17:30-17:50 - A User-centric Evaluation Methodology for Informed Provisioning of Supercomputing Resources in Academic Institutions
17:50-18:10 - EfiMon: A Process Analyser for Granular Power Consumption Prediction
18:10-18:30 - Leveraging CPU-FPGA Co-design for Matrix Profile Computation
16:50-17:10 - Enhancing Reverse Time Migration Simulations in HPC Systems through I/O and Computation Overlapping
17:10-17:30 - MKL Vs cuBLAS
17:30-17:50 - A User-centric Evaluation Methodology for Informed Provisioning of Supercomputing Resources in Academic Institutions
17:50-18:10 - EfiMon: A Process Analyser for Granular Power Consumption Prediction
18:10-18:30 - Leveraging CPU-FPGA Co-design for Matrix Profile Computation
Main Courtyard, d’Etigny Auditorium
16:45 - 18:45
HPC-IA Track
16:30-16:50: Web system for recognizing actions of physical violence in urban spaces using CNN with Transfer Learning. Presented by Wilder Suarez Romero
16:50-17:10: Quantized SG-MCMC for Bayesian Deep Posterior Compression.
17:10-17:30: No Plankton Left Behind: Preliminary results on massive plankton image recognition. Presented by Juan Perez Sofía Callejas
17:30-17:50: Machine Learning Regression-based Prediction for Improving Performance and Energy Consumption in HPC platforms.
17:50-18:10: A new computational framework for crop yield estimation and phenological monitoring.
18:10-18:30: Enhancing Histopathological Image Classification through StyleGAN2-ADA-based Data Augmentation.
16:50-17:10: Quantized SG-MCMC for Bayesian Deep Posterior Compression.
17:10-17:30: No Plankton Left Behind: Preliminary results on massive plankton image recognition. Presented by Juan Perez Sofía Callejas
17:30-17:50: Machine Learning Regression-based Prediction for Improving Performance and Energy Consumption in HPC platforms.
17:50-18:10: A new computational framework for crop yield estimation and phenological monitoring.
18:10-18:30: Enhancing Histopathological Image Classification through StyleGAN2-ADA-based Data Augmentation.
North Building, 7th Floor, John Von Neumann Room
16:45 - 18:45
HPC-Aplications Track
16:30-16:50 - Strategies to Reduce Memory Consumption in Software Quantum Computing Simulators
16:50-17:10 - Adaptive Edge-Based AIoT Architecture for Efficient Retraining and Sustainable Monitoring of Ephemeral Streams
17:10-17:30 - Multi-GPU Tomographic Reconstructions of Large Volumes in the Frequency Domain
17:30-17:50 - Accelerating tomographic artifact removal using a multi-GPU system
17:50-18:10 - A parallel multi-threading global energy balance for a room thermal analysis in an unsteady state
18:10-18:30 - Parallel Computing Strategies in WRF: The Role of MPI, OpenMP, & NUMA Affinity
16:50-17:10 - Adaptive Edge-Based AIoT Architecture for Efficient Retraining and Sustainable Monitoring of Ephemeral Streams
17:10-17:30 - Multi-GPU Tomographic Reconstructions of Large Volumes in the Frequency Domain
17:30-17:50 - Accelerating tomographic artifact removal using a multi-GPU system
17:50-18:10 - A parallel multi-threading global energy balance for a room thermal analysis in an unsteady state
18:10-18:30 - Parallel Computing Strategies in WRF: The Role of MPI, OpenMP, & NUMA Affinity
North Building, 3rd Floor, Picarte Auditorium
17:00 - 18:00
Aurora: HPC e IA a exaescala – Desafíos y aprendizaje (Sponsors: Intel)
Igor Freitas (Intel)
Thursday 03 October
08:30 - 18:00
Registration
Please remember to present the QR code you received via email.
Main Courtyard, d’Etigny Auditorium
09:00 - 09:45
Panel – Industry Snapshots: Sponsor Perspectives in HPC
This panel provides an exclusive platform for our Gold and Silver sponsors to share their key insigh ...
Main Courtyard, d’Etigny Auditorium
09:00 - 10:30
Simplificando la Complejidad: Software Científico como Servicio con Spack
Daniel Filipe Jana (SCITAS - École Polytechnique Fédérale de Lausanne - Suiza)
09:45 - 10:45
Keynote Talk – Marta Mattoso
Trusting data science workflows in HPC
Trusting outcomes from scientific workflows is related to the reproducibility and traceability of th ...
Main Courtyard, d’Etigny Auditorium
10:30 - 11:15
Sponsors: AMD
10:45 - 11:05
Industrial Talk Diamond – Lenovo
Main Courtyard, d’Etigny Auditorium
11:05 - 11:45
Coffee break, Posters and Expo
Main Courtyard, d’Etigny Auditorium
11:15 - 11:45
Coffee
11:45 - 12:45
Compilando tu primer stack: hands-on session
Daniel Filipe Jana (SCITAS - École Polytechnique Fédérale de Lausanne - Suiza)
11:45 - 12:25
Invited Talk – José María Cecilia
HPC and Data analytics to solve socio-ambiental issues
Coastal lagoons are ecosystems with important environmental and socioeconomic value. However, these ...
Main Courtyard, d’Etigny Auditorium
12:25 - 12:45
Industrial Talk Platinum – VersatusHPC+BeeGFS
Main Courtyard, d’Etigny Auditorium
12:45 - 14:30
Lunch Time
Av. Almte. Blanco Encalada 2540, Santiago, Región Metropolitana
Club House, Club Hípico, Santiago.
12:45 - 14:30
Almuerzo
14:30 - 15:00
Sponsors: Eviden
14:30 - 15:00
Special Talk – Ulises Cortés
When Sally Met Harry or When AI Met HPC
We will discuss the complementary nature of AI and HPC, illustrating how the combination enhances co ...
Main Courtyard, d’Etigny Auditorium
15:00 - 16:00
Benchmarking ¿Anda todo bien?
Edwin Fernando Posada (Oak Ridge National Laboratory – Estados Unidos de América)
15:00 - 16:00
Panel – Leadership Perspectives: The Role of HPC in Transforming LATAM
This unprecedented panel brings together top executives from AMD, NVIDIA, Intel, and RISC-V, marking ...
Main Courtyard, d’Etigny Auditorium
16:00 - 16:30
Coffee
16:00 - 16:30
Coffee break, Posters and Expo
Main Courtyard, d’Etigny Auditorium
16:30 - 17:30
HPC Track
16:30-16:50: Impact of job scheduling policy changes on user behaviour and system response: The case of the Santos Dumont supercomputer in Brazil. Presented by Antônio Tadeu Gomes
16:50-17:10: A Study of Performance Portability in Plasma Physics Simulations. Presented by Josef Ruzicka
17:10-17:30: High performance computing for auto supervised machine learning training: parallel-distributed implementation of the Word2Vec algorithm for training word embeddings.
16:50-17:10: A Study of Performance Portability in Plasma Physics Simulations. Presented by Josef Ruzicka
17:10-17:30: High performance computing for auto supervised machine learning training: parallel-distributed implementation of the Word2Vec algorithm for training word embeddings.
Main Courtyard, d’Etigny Auditorium
16:30 - 17:15
Sponsors: SONDA+IBM
Sponsors: SONDA+IBM
17:30 - 18:00
Mateo Valero Award Ceremony
Main Courtyard, d’Etigny Auditorium
19:30 - 22:00
Special Event and Gala Dinner
Av. Sta. María 1742, Providencia, Región Metropolitana
Salón Pedro de Valdivia, Hotel Sheraton, Providencia.
Friday 04 October
09:00 - 10:00
Keynote Talk – Bárbara Poblete
From fake-news to bias: Innovating through deep understanding of social data and diverse needs
In this talk I’ll provide an overview, based on firsthand experience, of difficult social (and tec ...
Main Courtyard, d’Etigny Auditorium
09:00 - 10:30
Sponsors: EMTEC, ENTEL
10:00 - 10:20
Panel – Funding Opportunities in Collaborative Supercomputing Infrastructure in LAC
This panel focuses on fostering collaboration and advancing High-Performance Computing (HPC) infrast ...
Main Courtyard, d’Etigny Auditorium
10:30 - 11:00
Casos de Estudio Latinoamericanos
Ginés Guerrero (NLHPC - Laboratorio Nacional de Computación de Alto Rendimiento - Chile),
Pablo Loyber (Servicio Meteorológico Nacional – Argentina),
Marcos Mazzini (Centro de Computación de Alto Desempeño de la Universidad Nacional de Cordoba - Argentina),
André Ramos Carneiro (Laboratório Nacional de Computação Científica - Brasil),
Sergio Neschmanow (Cluster UY – Universidad de la República – Uruguay),
Carlos Jaime Barrios Hernández (Universidad Industrial de Santander - Colombia),
Isidoro Gitler (ABACUS – Laboratorio de Matemática Aplicada y Cómputo de Alto Rendimiento – México)
Pablo Loyber (Servicio Meteorológico Nacional – Argentina),
Marcos Mazzini (Centro de Computación de Alto Desempeño de la Universidad Nacional de Cordoba - Argentina),
André Ramos Carneiro (Laboratório Nacional de Computação Científica - Brasil),
Sergio Neschmanow (Cluster UY – Universidad de la República – Uruguay),
Carlos Jaime Barrios Hernández (Universidad Industrial de Santander - Colombia),
Isidoro Gitler (ABACUS – Laboratorio de Matemática Aplicada y Cómputo de Alto Rendimiento – México)
11:00 - 11:40
Coffee
11:20 - 12:00
Coffee break, Posters and Expo
Enrique d'Etigny Auditorium Break Room - Below Auditorium
11:40 - 13:00
Case studies cont’d.
12:00 - 12:30
Special Talk – Carlos Jaime and Philippe Navaux
Towards Large Scale and Robust Cyberinfrastructures for Science and Technology in Latin America and the Caribbean
SCALAC is the Advanced Computing System for Latin America and the Caribbean, an alliance of infrastr ...
Main Courtyard, d’Etigny Auditorium
12:30 - 13:00
Invited Keynote – Edson Borin
Training AI Models on Massive Datasets with DASF
The exponential growth of data in the modern era presents significant challenges in training artific ...
Main Courtyard, d’Etigny Auditorium
13:00 - 13:30
Closing Remarks and Best Papers & Poster Prize
Main Courtyard, d’Etigny Auditorium
Wednesday 02 October
08:30 - 18:00
Registration
Please remember to present the QR code you received via email.
Main Courtyard, d’Etigny Auditorium
09:00 - 09:45
Welcome
Main Courtyard, d’Etigny Auditorium
09:45 - 10:45
Keynote Talk – Bernd Mohr
Parallel Performance Analysis at Scale: From Single Node to one Million HPC Cores
Current high-end HPC systems consist of complex configurations of potentially heterogeneous componen ...
Main Courtyard, d’Etigny Auditorium
10:45 - 11:05
Industrial Talk Diamond – Dell+AMD+NVIDIA
Main Courtyard, d’Etigny Auditorium
11:05 - 11:45
Coffee Break and Industrial Expo
Enrique d'Etigny Auditorium Break Room - Below Auditorium
11:45 - 12:25
Invited Talk – Alberto Ros
Timely Prefetching for High Performance Computers
Prefetching instructions and data is a fundamental technique for designing high-performance computer ...
Main Courtyard, d’Etigny Auditorium
12:25 - 12:45
Industrial Talk Platinum – Intel
Main Courtyard, d’Etigny Auditorium
12:45 - 14:15
Lunch Time
Av. Almte. Blanco Encalada 2540, Santiago, Región Metropolitana
Club House, Club Hípico, Santiago.
14:15 - 15:15
Special Talk – Mateo Valero
Latest Trends in High Performance Computer Architectures
In 1956, the term Artificial Intelligence was coined. From then on, research began that was thought ...
15:15 - 16:15
Panel – Government and Public Relations
The main objective of the panel is to understand the strategies and plans of different governments i ...
Main Courtyard, d’Etigny Auditorium
16:15 - 16:45
Poster Session – Opening
16:15 - 16:45
Coffee break, Posters and Expo
16:45 - 18:45
HPC Track
16:30-16:50 - A Comprehensive Analysis of Process Energy Consumption on Multi-Socket Systems with GPUs
16:50-17:10 - Enhancing Reverse Time Migration Simulations in HPC Systems through I/O and Computation Overlapping
17:10-17:30 - MKL Vs cuBLAS
17:30-17:50 - A User-centric Evaluation Methodology for Informed Provisioning of Supercomputing Resources in Academic Institutions
17:50-18:10 - EfiMon: A Process Analyser for Granular Power Consumption Prediction
18:10-18:30 - Leveraging CPU-FPGA Co-design for Matrix Profile Computation
16:50-17:10 - Enhancing Reverse Time Migration Simulations in HPC Systems through I/O and Computation Overlapping
17:10-17:30 - MKL Vs cuBLAS
17:30-17:50 - A User-centric Evaluation Methodology for Informed Provisioning of Supercomputing Resources in Academic Institutions
17:50-18:10 - EfiMon: A Process Analyser for Granular Power Consumption Prediction
18:10-18:30 - Leveraging CPU-FPGA Co-design for Matrix Profile Computation
Main Courtyard, d’Etigny Auditorium
16:45 - 18:45
HPC-IA Track
16:30-16:50: Web system for recognizing actions of physical violence in urban spaces using CNN with Transfer Learning. Presented by Wilder Suarez Romero
16:50-17:10: Quantized SG-MCMC for Bayesian Deep Posterior Compression.
17:10-17:30: No Plankton Left Behind: Preliminary results on massive plankton image recognition. Presented by Juan Perez Sofía Callejas
17:30-17:50: Machine Learning Regression-based Prediction for Improving Performance and Energy Consumption in HPC platforms.
17:50-18:10: A new computational framework for crop yield estimation and phenological monitoring.
18:10-18:30: Enhancing Histopathological Image Classification through StyleGAN2-ADA-based Data Augmentation.
16:50-17:10: Quantized SG-MCMC for Bayesian Deep Posterior Compression.
17:10-17:30: No Plankton Left Behind: Preliminary results on massive plankton image recognition. Presented by Juan Perez Sofía Callejas
17:30-17:50: Machine Learning Regression-based Prediction for Improving Performance and Energy Consumption in HPC platforms.
17:50-18:10: A new computational framework for crop yield estimation and phenological monitoring.
18:10-18:30: Enhancing Histopathological Image Classification through StyleGAN2-ADA-based Data Augmentation.
North Building, 7th Floor, John Von Neumann Room
16:45 - 18:45
HPC-Aplications Track
16:30-16:50 - Strategies to Reduce Memory Consumption in Software Quantum Computing Simulators
16:50-17:10 - Adaptive Edge-Based AIoT Architecture for Efficient Retraining and Sustainable Monitoring of Ephemeral Streams
17:10-17:30 - Multi-GPU Tomographic Reconstructions of Large Volumes in the Frequency Domain
17:30-17:50 - Accelerating tomographic artifact removal using a multi-GPU system
17:50-18:10 - A parallel multi-threading global energy balance for a room thermal analysis in an unsteady state
18:10-18:30 - Parallel Computing Strategies in WRF: The Role of MPI, OpenMP, & NUMA Affinity
16:50-17:10 - Adaptive Edge-Based AIoT Architecture for Efficient Retraining and Sustainable Monitoring of Ephemeral Streams
17:10-17:30 - Multi-GPU Tomographic Reconstructions of Large Volumes in the Frequency Domain
17:30-17:50 - Accelerating tomographic artifact removal using a multi-GPU system
17:50-18:10 - A parallel multi-threading global energy balance for a room thermal analysis in an unsteady state
18:10-18:30 - Parallel Computing Strategies in WRF: The Role of MPI, OpenMP, & NUMA Affinity
North Building, 3rd Floor, Picarte Auditorium
HPC Track
Enrique d'Etigny Auditorium
There are currently no events.
HPC-IA Track
Sala John Von Neumann
There are currently no events.
HPC+Aplications Track
B06 Room
There are currently no events.
Thursday 03 October
08:30 - 18:00
Registration
Please remember to present the QR code you received via email.
Main Courtyard, d’Etigny Auditorium
09:00 - 09:45
Panel – Industry Snapshots: Sponsor Perspectives in HPC
This panel provides an exclusive platform for our Gold and Silver sponsors to share their key insigh ...
Main Courtyard, d’Etigny Auditorium
09:45 - 10:45
Keynote Talk – Marta Mattoso
Trusting data science workflows in HPC
Trusting outcomes from scientific workflows is related to the reproducibility and traceability of th ...
Main Courtyard, d’Etigny Auditorium
10:45 - 11:05
Industrial Talk Diamond – Lenovo
Main Courtyard, d’Etigny Auditorium
11:05 - 11:45
Coffee break, Posters and Expo
Main Courtyard, d’Etigny Auditorium
11:45 - 12:25
Invited Talk – José María Cecilia
HPC and Data analytics to solve socio-ambiental issues
Coastal lagoons are ecosystems with important environmental and socioeconomic value. However, these ...
Main Courtyard, d’Etigny Auditorium
12:25 - 12:45
Industrial Talk Platinum – VersatusHPC+BeeGFS
Main Courtyard, d’Etigny Auditorium
12:45 - 14:30
Lunch Time
Av. Almte. Blanco Encalada 2540, Santiago, Región Metropolitana
Club House, Club Hípico, Santiago.
14:30 - 15:00
Special Talk – Ulises Cortés
When Sally Met Harry or When AI Met HPC
We will discuss the complementary nature of AI and HPC, illustrating how the combination enhances co ...
Main Courtyard, d’Etigny Auditorium
15:00 - 16:00
Panel – Leadership Perspectives: The Role of HPC in Transforming LATAM
This unprecedented panel brings together top executives from AMD, NVIDIA, Intel, and RISC-V, marking ...
Main Courtyard, d’Etigny Auditorium
16:00 - 16:30
Coffee break, Posters and Expo
Main Courtyard, d’Etigny Auditorium
16:30 - 17:30
HPC Track
16:30-16:50: Impact of job scheduling policy changes on user behaviour and system response: The case of the Santos Dumont supercomputer in Brazil. Presented by Antônio Tadeu Gomes
16:50-17:10: A Study of Performance Portability in Plasma Physics Simulations. Presented by Josef Ruzicka
17:10-17:30: High performance computing for auto supervised machine learning training: parallel-distributed implementation of the Word2Vec algorithm for training word embeddings.
16:50-17:10: A Study of Performance Portability in Plasma Physics Simulations. Presented by Josef Ruzicka
17:10-17:30: High performance computing for auto supervised machine learning training: parallel-distributed implementation of the Word2Vec algorithm for training word embeddings.
Main Courtyard, d’Etigny Auditorium
17:30 - 18:00
Mateo Valero Award Ceremony
Main Courtyard, d’Etigny Auditorium
19:30 - 22:00
Special Event and Gala Dinner
Av. Sta. María 1742, Providencia, Región Metropolitana
Salón Pedro de Valdivia, Hotel Sheraton, Providencia.
Friday 04 October
09:00 - 10:00
Keynote Talk – Bárbara Poblete
From fake-news to bias: Innovating through deep understanding of social data and diverse needs
In this talk I’ll provide an overview, based on firsthand experience, of difficult social (and tec ...
Main Courtyard, d’Etigny Auditorium
10:00 - 10:20
Panel – Funding Opportunities in Collaborative Supercomputing Infrastructure in LAC
This panel focuses on fostering collaboration and advancing High-Performance Computing (HPC) infrast ...
Main Courtyard, d’Etigny Auditorium
11:20 - 12:00
Coffee break, Posters and Expo
Enrique d'Etigny Auditorium Break Room - Below Auditorium
12:00 - 12:30
Special Talk – Carlos Jaime and Philippe Navaux
Towards Large Scale and Robust Cyberinfrastructures for Science and Technology in Latin America and the Caribbean
SCALAC is the Advanced Computing System for Latin America and the Caribbean, an alliance of infrastr ...
Main Courtyard, d’Etigny Auditorium
12:30 - 13:00
Invited Keynote – Edson Borin
Training AI Models on Massive Datasets with DASF
The exponential growth of data in the modern era presents significant challenges in training artific ...
Main Courtyard, d’Etigny Auditorium
13:00 - 13:30
Closing Remarks and Best Papers & Poster Prize
Main Courtyard, d’Etigny Auditorium