Program Overview

Program Overview

The Distributed and Parallel Computing (M2 DiPaC) program is a master's program that focuses on advanced topics in high-performance computing (HPC) and distributed algorithms for large-scale systems. The program is designed to provide students with the knowledge and skills necessary to design fast, scalable, and robust solutions for applications in AI, big data analytics, scientific simulations, and quantum-enabled workflows.

Objectives

The program's objectives are to:

• Provide students with a solid foundation in parallel programming models and performance engineering on modern supercomputers and accelerators.
• Equip students with the knowledge and skills necessary to develop and analyze scalable, robust algorithms with theoretical guarantees on scalability, consensus, termination, and fault tolerance.
• Enable students to optimize HPC code across the stack, including complexity, memory locality, vectorization, accelerator use, communication, I/O, and networks.

Skills

The program aims to provide students with the following skills:

• The ability to read and understand research articles in the fields of distributed, parallel, and quantum computing.
• The ability to design and develop high-quality parallel and distributed algorithms and software that meet latency, throughput, and performance targets on target HPC architectures.
• The ability to develop and analyze scalable, robust algorithms with theoretical guarantees on scalability, consensus, termination, and fault tolerance.

Career Opportunities

The program provides students with a range of career opportunities, including:

• Researcher or lecturer
• Engineer (R&D, control, production)
• Data scientist
• AI specialist
• R&D project manager
• Consultant
• Data protection officer
• Information systems manager

Further Study Opportunities

The program also provides students with the opportunity to pursue further study, including:

• Doctorate
• Engineering school
• Research internship
• Thesis

Fees and Scholarships

The program's tuition fees vary depending on the student's personal circumstances. A limited number of scholarships are available for exceptional candidates.

Admission

The program's admission criteria include:

• A solid foundation in computer science with a background in parallel programming and distributed systems.
• Strong mathematical skills, particularly in linear algebra.
• A motivation letter and CV.
• Transcripts of all courses since high school.
• A letter of recommendation or internship evaluation (optional).

Application Period

The application period for the program is from April 15th to May 30th.

Location

The program is located in Orsay and Gif-sur-Yvette.

Academic Partners

The program has academic partnerships with several institutions, including:

• École Polytechnique
• Télécom Paris
• INRIA
• Sorbonne Université
• Université de Paris
• Technion - Israel Institute of Technology
• University of Tennessee
• Old Dominion University
• École Polytechnique Fédérale de Lausanne
• Lisbon University
• Karlsruhe Institute of Technology
• University of Vienna

Program Structure

The program includes:

• Seven core disciplinary courses in high-performance, parallel, and distributed computing.
• Two elective courses to further specialize in high-performance data analysis and AI (HPDA) or hybrid HPC/quantum computing (HQI).
• One soft-skill course to reinforce transferable professional skills.
• A mandatory 6-month internship on M2 DiPaC-related themes.

Resources and Practice

The program provides students with access to university clusters and partner supercomputers for hands-on labs, course projects, code development, and tuning. Students also have access to open-source toolchains and libraries widely adopted by the HPC community.