HCMUT HPC Summer School 2026

Program

A 3-day learning journey from HPC foundations to scalable AI and data-intensive workloads.

Program Overview

The program is designed as a progression from foundational concepts to hands-on practice and applied exploration. Students begin with the core ideas of HPC systems and cluster execution, then move toward modern AI and data-intensive workloads, and finally apply what they have learned through a mini challenge, seminar, and closing activities.

Day 1

Foundations of HPC: From Local Code to Cluster Execution

Morning

Opening Ceremony and Introduction to High Performance Computing
Welcome remarks, school orientation, and a first look at why HPC matters for AI, Big Data, scientific computing, and shared research infrastructure.
Parallel Computing Fundamentals and First Parallel Programs
Core ideas behind parallel execution, performance bottlenecks, scalability, and a guided first step into simple parallel programming practice.

Afternoon

Working with HPC Environments
Practical orientation to Linux-based cluster workflows, remote access, files, software environments, modules, and reproducible execution habits.
Slurm Job Submission and Cluster Practice
Hands-on practice with job scripts, resource requests, queue behavior, monitoring, and running basic workloads on shared infrastructure.

Day 2

Modern HPC for AI and Data-Intensive Workloads

Morning

Distributed Data Processing Concepts
A conceptual tour of data partitioning, pipeline decomposition, map-reduce style thinking, communication overhead, and workload coordination.
Hands-on Lab: Scalable Data Processing Workflow
Guided implementation of a small distributed processing workflow, with attention to workload splitting, execution control, and result aggregation.

Afternoon

Containerization for HPC Workloads
How containers support reproducible environments, dependency management, and portable application execution on shared HPC systems.
Scaling AI/LLM Workloads on HPC
Practical patterns for moving AI and LLM-related workloads beyond one machine through orchestration, monitoring, and scalable execution.

Day 3

Mini Challenge, Research Community, and Future Directions

Morning

Mini Challenge for HPC, Big Data, and AI
A team-based applied activity where students use selected school concepts to design, run, and explain a scalable workflow or experiment.

Afternoon

Seminar, Panel Discussion, and Closing Ceremony
Invited sharing, community discussion, participant reflection, recognition, and next-step orientation for continuing in HPC-related pathways.

The agenda is subject to change. Detailed session information and preparation instructions will be shared with accepted students.

Learning Outcomes

By the end of the school, students are expected to:

Understand the basic concepts of HPC systems and cluster environments.
Explain why modern AI, data, and scientific workloads require scalable computing resources.
Run basic jobs in an HPC environment.
Understand the role of schedulers, containers, and monitoring in HPC workflows.
Explore scalable data and AI workload patterns.
Work with peers in a short technical challenge.
Communicate technical results and reflect on future research or engineering directions.

Ready to start local and compute at scale?

Join HCMUT HPC Summer School 2026 and take your first step into scalable computing, AI systems, and research-oriented engineering. Contact bdc+hpcschool@hcmut.edu.vn for questions.

Apply Now

↑

Program

Program Overview

Foundations of HPC: From Local Code to Cluster Execution

Modern HPC for AI and Data-Intensive Workloads

Mini Challenge, Research Community, and Future Directions

Learning Outcomes

Ready to start local and compute at scale?

Ready to compute at scale?