2022-2023 / INFO9012-1

Parallel Programming

Duration

25h Th, 25h Pr

Number of credits

 Bachelor of Science (BSc) in Engineering5 crédits 
 Bachelor of Science (BSc) in Computer Science5 crédits 
 Master of Science (MSc) in Data Science5 crédits 
 Master of Science (MSc) in Computer Science and Engineering5 crédits 
 Master of Science (MSc) in Data Science and Engineering5 crédits 
 Master of Science (MSc) in Computer Science5 crédits 

Lecturer

Pascal Fontaine

Language(s) of instruction

English language

Organisation and examination

Teaching in the second semester

Schedule

Schedule online

Units courses prerequisite and corequisite

Prerequisite or corequisite units are presented within each program

Learning unit contents

This course addresses some practical and theoretical aspects of distributed computing and parallel programming. It will notably introduce parallel programming models, the notions of process and thread, race conditions, mutual exclusion, memory models, semaphores, monitors, and conditions.  Some classical problems are studied.  On the more practical side, the course will be based on examples in classical programming languages (C, C++, Java, Python) and will quickly introduce OpenMP.  Finally, message passing will be explained, MPI and Go being used to illustrate the practice of message passing and queues.

 

Learning outcomes of the learning unit

At the end of this course, the student will be aware of the pitfalls of parallel and distributed programming.  She/he will know the basic algorithms to solve typical issues that occur in distributed systems, and will be able to apply them in concrete applications.

Prerequisite knowledge and skills

Excellent familiarity with sequential imperative programming is mandatory.

Examples are provided in C, C++, Java and Python.  We assume the student to be sufficiently at ease with programming not to be disturbed if some notions are explained in a less well-known language.

The project will be done in C++.  Again, since most of the code is provided, and we do not require from the student to write a large amount of code, there is no formal pre-requisite for C++.  However, again, the student must be sufficiently familiar with imperative programming to get along writing a few hundred lines of C++ code.

Students must have access to a machine under Linux, with a working X11 server.  A Raspberry Pi or a well-installed virtual machine is enough.  Machines are available at the university (ms8xx.montefiore.ulg.ac.be)

Planned learning activities and teaching methods

The course comprises:

  • ex cathedra courses, that introduce the concepts and techniques
  • problem sessions
  • hands-on sessions in the lab (C / go), to get practical knowledge of parallel programming. 
a more ambitious homework project (C++)

Mode of delivery (face to face, distance learning, hybrid learning)

Face-to-face course


Additional information:

Lectures are given in English.  2nd quadrimester, face-to-face.

Recommended or required readings

Reference text, notes and transparencies available on the course space on eCampus.

Exam(s) in session

Any session

- In-person

written exam ( open-ended questions )

Written work / report


Additional information:

Grading of programming assignments (30%), written exam (no oral exam).

Work placement(s)

Organizational remarks

The contents of the theoretical and practical sessions, as well as the assignments and useful links, will be made available on the e-Campus space for the course.

Contacts

Teacher: Pascal Fontaine
Phone: 04 366 28 75
e-mail: Pascal.Fontaine@uliege.be

Association of one or more MOOCs