High-Performance-Computing

This repository contains the activities required during the subject High Perfomance Computing of the University of Lleida Master's Degree in Informatics Engineering.

This is a repository for PRACTISING, INSPIRATION and to review the solutions of a previous year, as they should not be the same.

Give me a star ⭐ if you find it useful. You can contact me if any question.

Thanks

Many thanks to my colleague Francesc Contreras for his contribution to the activities of the course.

Learning Objectives

• Knowledge the use of benchmarking tools applied to the high performance computing environments.
• Capacity for analysis and understanding the computer perfomance.
• Knowledge of the parallel programming paradigms.
• Implementation and debugging of parallel applications using the OpenMP and MPI paradigms.
• Knowledge and understanding of the cloud computing paradigm.
• Practical use of frameworks and cloud computing technologies.
• Knowledge and understanding the architecture and operation for the most common HPC Infrastructures most
• booming in the market.
• Solving numerical algorithms using a Hybridization of OpenMP and MPI technologies and evaluating its performance.
• Communication of ideas and concepts in english in an understandable way, writing and speaking.

Competences

University of Lleida (UdL) strategic competences

• UdL1: Appropriate skills in oral and written language.
• UdL2: Command of a foreign language.

Higher Polytecnic School (EPS) cross-disciplinary competences

• EPS3: Capacity to convey information, ideas, problems and solutions to both a specialized and no specialized public.
• EPS4: Capacity to conceive, design and implement projects and/or contribute to new solutions, using engineering tools.

General comptences

• CG4: Capacity for mathematical modeling, calculation and simulation in technology and engineering company centers, particularly in research, development and innovation tasks in all areas related to informatic engineering.
• CG8: Capacity to apply the knowledge acquired for solving problems in new and unfamiliar situations within broader and more multidisciplinary contexts, and to be capable of integrating this knowledge.

Basic competences

• CB1: Possess knowledge and understanding that provide a basis or opportunity for originality in developing and / or applying ideas, often in a research context.
• CB2: That the students can apply their knowledge and their ability to solve problems in new or unfamiliar environments within broader (or multidisciplinary) contexts related to their field of study.

Degree-specific competences

• CE10: Capacity to understand and apply advanced knowledge in high-performance computing and numerical or computational methods to problems of engineering.
• CE12: Capacity to apply mathematical, statistical and artificial intelligence methods, design and develop applications, services, intelligent systems and systems based on knowledge.

Subject Contents

1. Introduction: High Performance Computing (2h F+4h NF)

2. Introduction to the parallel processing and Benchmarking(4h F+12h NF)

   • Introduction
   • Concepts
   • Parallel programming models
   • Design of parallel applications
   • Limits of the parallel processing
   • Benchmarking Tools

3. Shared-memory Parallel Programming with OpenMP (8h F+20h NF)

   • Main Characteristics
   • Parallel Regions
   • Data Scoping
   • OpenMP worksharing for loops
   • Reductions
   • Loop Scheduling
   • Synchronization

4. Distributed-memory Parallel Programming with MPI (10h F+30h NF)

   • Main Characteristics
   • Messages and Point-to-point communications
   • Collevtive communication
   • Grouping Data for communication
   • Communicators and Topologies

5. Cloud Computing (4h F+10h NF)

   • Cloud Essentials
   • Anatomy of the cloud
   • Oportunities
   • OpenNebula

6. Parallel computers (2h F+6h NF)

   • Types of parallel computers
   • Analysis of main supercomputers located in TOP500

References

• P.S. Pacheco (1997). Parallel Programming with MPI. Morgan Kaufmann Publishers.
• P. Pacheco (2011). An Introduction to Parallel Programming. Morgan Kaufmann Publishers.
• R. Chandra, L. Dagum, D. Kohr (2001). Parallel Programming in OpenMP. Morgan Kaufmann Publishers.