Advanced HPC (High Performance Computing) Practical Training

Education title: Advanced HPC (High Performance Computing) Practical Training
Name of the relevant MontEDIH service: Advanced HPC hands-on training

Service provider: University of Donja Gorica – MontEDIH Service

Trainer: Prof. dr Ramo Šendelj; Prof.dr Ivana Ognjanović; Mr Luka Laković

Target users: Small and medium-sized enterprises (SMEs) and public institutions aiming to enhance their research and business capacities through advanced application of HPC resources, workflow optimization, large dataset analysis, and the implementation of AI/ML solutions.

 

Training Content

The training is an intensive three-day program that combines theoretical knowledge with practical work in an HPC environment. Participants will deepen their understanding of advanced HPC concepts, master parallelization and resource optimization, and learn to develop and evaluate complex workflows and Proof-of-Concept (PoC) projects.

 

Day 1 focuses on advanced data preparation and the HPC environment. Participants will learn to plan and map HPC workflows, manage and optimize datasets, prepare data for parallel processing, and configure flexible software environments and containers.

Day 2 is dedicated to execution and optimization. Participants will master advanced HPC job management, application performance optimization, parallel programming, and resource monitoring. A special focus is placed on the efficient use of CPU, GPU, and memory, as well as identifying and resolving bottlenecks in HPC workflows.

 

Day 3 covers the application of HPC in AI and advanced analytics, the development of Proof-of-Concept solutions, and the evaluation of HPC return on investment (ROI). The final workshop integrates all acquired skills through a real-world end-to-end project, ranging from data preparation to results analysis.

 

Learning Outcomes

Upon completion of the training, participants will be able to:

• Design and implement complex HPC workflows.
• Efficiently manage and optimize large datasets.
• Configure and utilize software environments and containers.
• Execute, optimize, and parallelize applications on HPC systems.
• Apply parallel models (MPI, OpenMP, GPU).
• Analyze performance and identify bottlenecks.
• Develop a functional Proof-of-Concept using HPC.
• Assess the value and cost-effectiveness of HPC solutions.
• Independently apply HPC in business and research contexts.

 

Schedule and Organization

• Duration: 3 days × 4 sessions (60 minutes per session)
• Format: Theoretical-practical sessions, exercises in a live HPC environment, discussions, and a final workshop.
• Price: Free of charge for MontEDIH service users.
• Date and Time: To be determined in coordination with registered users upon group formation.

 

Additional notes:

• The training is tailored to the needs of SMEs and the public sector, with a focus on the advanced application of HPC in data analysis, AI/ML projects, and workflow optimization.
• Participants will receive instructor support throughout all practical sessions and the final workshop.
• Upon completion, participants will have gained the competencies required for the independent planning, implementation, and evaluation of HPC solutions in business and research scenarios.

Download Agenda