Crèdits
3
Tipus
Optativa
Requisits
Aquesta assignatura no té requisits
, però té capacitats prèvies
Departament
AC
Web
https://torres.ai/HPC4AI-MEI
Mail
jordi.torres@upc.edu
Rather than treating deep learning frameworks and tools as black boxes, the course adopts a system-oriented perspective. It guides students through the complete execution workflow of AI training, from hardware architecture and system software to job scheduling, parallel execution, performance measurement, and scalability analysis. The emphasis is on execution behavior: how computation, memory, communication, and coordination interact, and how these interactions determine performance, efficiency, and cost.
A central premise of the course is that the nature of engineering work in AI is changing. Modern AI tools can generate training scripts, pipelines, and even distributed execution logic with minimal effort. As a result, writing code is no longer the primary challenge. The real difficulty, and the real value, lies in understanding whether that code scales, where bottlenecks appear, when efficiency is lost, and what trade-offs are being made when more resources are used.
For this reason, the course explicitly allows and acknowledges the use of modern AI tools (such as code assistants, agentic systems, or automated code generators). However, the course is not about code authorship or syntax. It is about developing the ability to reason about performance, scalability, efficiency, and cost when training deep learning models on real HPC systems. Students are expected to understand what is being executed, how it behaves at scale, and why performance changes as observed.
Hands-on experimentation is a core component of the course. Through a sequence of laboratory activities, students train deep learning models using single and multiple GPUs, explore parallel and distributed training strategies, and analyze scalability and performance behavior under realistic conditions. All laboratory work and assessments are evaluated based on the quality of experimental setup, the relevance of performance measurements, the interpretation of results, and the soundness of scalability and cost¿benefit reasoning.
The course material is self-contained and based on the official course textbook, which serves as the main reference for both theoretical concepts and practical activities. No prior experience with supercomputers is required, and deep learning concepts are introduced progressively as needed.
Ultimately, HPC4AI is not a course about recipes or fixed solutions. It is a course about developing engineering judgment. As code generation becomes cheaper and more accessible, the ability to measure, reason, and decide becomes essential. This course is designed to develop precisely that ability.
Details specific to the 2026 edition of the course can be found on the course web page:
https://torres.ai/HPC4AI-MEI
Professorat
Responsable
- Jordi Torres Viñals ( torres@ac.upc.edu )
Hores setmanals
Teoria
2
Problemes
0
Laboratori
2
Aprenentatge dirigit
0
Aprenentatge autònom
7.1
Competències
Direcció i gestió
Específiques
Genèriques
Actitud adequada davant el treball
Bàsiques
Objectius
-
OE1: Foundations of HPC platforms for AI: comprendre l'arquitectura, els components principals i l'entorn software d'una plataforma de supercomputació moderna orientada a càrregues de treball d'intel·ligència artificial.
Competències relacionades: CTE6, CG1, CG6, CG7, CG8, -
OE2: Practical use of a supercomputer for AI workloads: adquirir autonomia bàsica en l'ús d'un supercomputador real, incloent accés, gestió de recursos i execució de treballs per a aplicacions d'intel·ligència artificial.
Competències relacionades: CTE6, CG1, CB6, CG8, -
OE3: Fundamentals of Deep Learning for HPC users: entendre els principis fonamentals del Deep Learning necessaris per entrenar models en entorns de supercomputació, sense requerir coneixements previs avançats.
Competències relacionades: CTE9, CG4, CG8, -
OE4: Parallel training of Deep Learning models: comprendre i aplicar tècniques d¿entrenament paral·lel de models de Deep Learning utilitzant múltiples GPUs en un o diversos nodes (servidors) de computació
Competències relacionades: CTE6, CTE9, CG1, CB6, CB9, -
OE5: Performance analysis and optimization of AI training: analitzar el rendiment de l¿entrenament de models d'intel·ligència artificial mitjançant mètriques com throughput, speedup i eficiència, i aplicar tècniques bàsiques d¿optimització.
Competències relacionades: CTE6, CTE9, CG1, CG4, -
OE6: Experimental evaluation and communication of results: avaluar experimentalment resultats obtinguts en un entorn de supercomputació i comunicar conclusions tècniques de manera clara, estructurada i argumentada.
Competències relacionades: CDG1, CTR5, CB8, CB9,
Continguts
-
C1: HPC platforms and software ecosystem for AI
Arquitectura d'un supercomputador modern, components hardware, sistema operatiu, entorn de programari i stack bàsic de software per a càrregues de treball d'intel·ligència artificial. -
C2: Accessing and using a supercomputer for AI workloads
Accés a un supercomputador, gestió de comptes d'usuari, sistemes de cues, SLURM i execució de treballs per a aplicacions de Deep Learning. -
C3: Deep Learning fundamentals for HPC environments
Conceptes bàsics de Deep Learning necessaris per entrenar models en entorns HPC: xarxes neuronals, entrenament, datasets i fluxos de treball (no podem suposar coneixements prèvis). -
C4: Parallel training of Deep Learning models
Entrenament paral·lel de models de Deep Learning utilitzant múltiples GPUs, incloent estratègies de paral·lelisme i frameworks de programació. -
C5: Performance metrics and optimization of AI training
Anàlisi del rendiment de l'entrenament de models d'IA mitjançant mètriques com throughput, speedup i eficiència, i tècniques bàsiques d'optimització. -
C6: Experimental evaluation and presentation of results
Avaluació experimental de resultats obtinguts en un entorn HPC i comunicació clara de conclusions mitjançant informes i presentacions tècniques.
Activitats
Activitat Acte avaluatiu
Teoria
1h
Problemes
0h
Laboratori
0h
Aprenentatge dirigit
0h
Aprenentatge autònom
1h
Teoria
2.5h
Problemes
0h
Laboratori
0h
Aprenentatge dirigit
0h
Aprenentatge autònom
2h
Teoria
2h
Problemes
0h
Laboratori
1.5h
Aprenentatge dirigit
0h
Aprenentatge autònom
3h
Teoria
3h
Problemes
0h
Laboratori
1h
Aprenentatge dirigit
0h
Aprenentatge autònom
4.5h
Teoria
2.5h
Problemes
0h
Laboratori
2h
Aprenentatge dirigit
0h
Aprenentatge autònom
5h
Teoria
0h
Problemes
0h
Laboratori
0h
Aprenentatge dirigit
0h
Aprenentatge autònom
9h
Teoria
0h
Problemes
0h
Laboratori
0h
Aprenentatge dirigit
0h
Aprenentatge autònom
2h
Teoria
0h
Problemes
0h
Laboratori
0h
Aprenentatge dirigit
0h
Aprenentatge autònom
0.4h
Metodologia docent
The course follows an active learning and continuous assessment approach, combining theoretical lectures, hands-on laboratory work, autonomous learning, and student presentations.Theoretical sessions are delivered through participatory lectures, where the instructor introduces the fundamental concepts related to high-performance computing platforms, deep learning fundamentals, parallel training strategies, and performance analysis for artificial intelligence workloads. Students are expected to actively participate in discussions during these sessions.
Hands-on activities constitute a central component of the course and are based on a learn-by-doing methodology. These activities focus on practical experimentation using a real supercomputing environment (MareNostrum 5). Part of the hands-on work is carried out during regular class sessions, while the remaining work is completed outside the classroom as autonomous learning. All hands-on activities require the submission of corresponding reports and, in some cases, technical presentations through the institutional learning platform (Racó).
Autonomous learning is mainly based on the detailed study of the course textbook, which constitutes the main reference material for the subject. Students are also required to prepare presentations and technical material related to their practical work.
Student presentations play an important role in the course. Individual students or groups are randomly selected to present their work and results in class. Peer evaluation is incorporated as part of the learning process, encouraging critical analysis and constructive feedback.
Regular attendance and active participation are expected. Students are responsible for all material covered in class, including announcements, assignments, and project guidelines, regardless of attendance. It is the student¿s responsibility to obtain any missed material.
Mètode d'avaluació
The evaluation of this course is based on a continuous assessment system, strongly focused on practical work and active participation.The final grade is composed of the following elements:
- Attendance and participation: 20%
Regular attendance and active participation in lectures, discussions, and hands-on sessions.
Attendance is mandatory. To qualify for continuous assessment, students must attend at least 80% of the class sessions.
- Hands-on activities (laboratory work): 60%
Evaluation of the practical laboratory activities carried out throughout the course (LAB 0 to LAB 4).
The instructor will assess the submitted work using a rubric that considers correctness, completeness, experimental results, and technical understanding.
Some students or groups will be randomly selected during the course to present and explain their laboratory work (LAB 0 to LAB 2). This mechanism is intended to ensure that all students prepare and understand their work thoroughly.
- Technical presentations and peer evaluation: 20%
During the final session of the course, all students will present either LAB 3 or LAB 4 (assigned randomly).
Presentations will be evaluated by the instructor and through peer evaluation, which will contribute to the final presentation grade.
Attendance on the presentation day is mandatory. Students who do not attend this session will not receive the presentation grade.
Requirements for continuous assessment: To qualify for continuous assessment, students must meet all the following requirements:
- Attendance: at least 80% of the class sessions.
- Hands-on activities: completion of at least 50% of the laboratory work.
Final exam option
- Students who do not meet the requirements for continuous assessment will have the option to take a final exam.
- This exam will evaluate the entire course content, including theoretical concepts, practical knowledge, and autonomous learning material based on the course book and laboratory activities.
- The final exam will be announced during the course. No documentation (printed or digital) will be allowed during the exam.
Bibliografia
Bàsic
-
Supercomputing for Artificial Intelligence: Foundations, Architectures, and Scaling Deep Learning
- Torres, Jordi,
WATCH THIS SPACE Book Series - Barcelona. Amazon KDP,
2025.
ISBN: 979-831932835-9
-
Slides of the course
- Torres, J,