Course syllabus

Principles of Concurrent Programming | TDA384/DIT392 (Jan 2024 - Mar 2024) 2023-2024

Course-PM

TDA384/DIT392 — Principles of Concurrent Programming (7,5hp)

Revised on 22 Aug 2024

Department of Computer Science and Engineering (CSE)

Quick links…

Course purpose

This course provides an introduction to the principles underlying concurrent systems, as well as to practical programming solutions to use concurrency in programs.

Concurrent and parallel programming has become ubiquitous in modern software and systems, where concurrency is leveraged to exploit physical parallelism and speed up computations, to provide interactive multi-tasking, and to handle the interactions with asynchronous external events. This course aims to provide an introduction to the principles underlying concurrent systems, as well as to practical programming solutions for modelling and exploiting concurrency in programs. Domains where such principles and practices are relevant include operating systems, distributed systems, real-time systems, and multicore architectures.

The official course syllabus is available on the Chalmers student portal and the GU website.

Schedule

For details on the lectures, see the course plan below and the lectures page. For a detailed schedule (including room information), see TimeEdit link.

Note: The course is running mostly on campus. Early morning labs and all demos will be held virtually.

Instructors and contact details

The primary means of communication for the course are:

This Canvas course room is the primary source of information about the course. All announcements will be made via Canvas.
The discussion forum is also hosted on Canvas . There you can ask questions about the course to the teachers and to other students. Sharing solutions to labs is forbidden — on Canvas or anywhere else.
Waglys is used for coordinating meeting with TAs during labs (see labs/links).
See information on lectures, labs, and exam.
The Teacher’s mailing list is pcp-teachers@lists.chalmers.se, which you shouldn’t use unless you really need to contact the teachers via email on a subject that is personal — please use the discussion forum to ask questions that are of general interest!
The Fire system where you can upload submissions to the labs. (see links).

Course responsibles

Lecturers/Examiners:

Nir Piterman (in LP1)

Gerardo Schneider (in LP3)

Teaching assistants (TAs)

The TAs are responsible for supervision of the lab sessions and grading.

See current TAs here.

Course literature

(If you find any broken links here, please report them to the instructors by email, so we can fix them!)

Course textbooks

The lectures themselves are the ultimate reference for the course topics. Nevertheless, we refer you to the two official course textbooks for additional details. The course’s list of lectures mentions which chapters and sections of the textbooks are covered in each lecture.

M. Ben-Ari: Principles of Concurrent and Distributed Programming, 2nd Edition, Addison-Wesley

The book has a website with downloadable slides and code examples
M. Herlihy & N. Shavit: The Art of Multiprocessor Programming, Morgan Kaufmann

The book is available online through the Chalmers library and the GU library

Additional reading

See the course literature page for additional resources.

Course plan

The course normally has two lectures per week. The lecture slides of each lecture are available on the course website and in Canvas latest the day before each lecture.

There are three supervised lab sessions per week. Students can attend any of the supervised lab session and receive help and tutoring from the supervising assistants on the lab assignments that are currently due. There are three labs, which require the students to give a brief demo to the teaching assistants during one of the designated lab sessions.

The course consists of the following lectures and labs.

Lectures

Introduction to concurrent programmin
Races locks and semaphores
Models of concurrency and synchronization algorithms
Models of concurrency and synchronization algorithms
Synchronization problems with semaphores
Synchronization problems with semaphores & Monitors
Parallelizing computations
Intro to FP in Erlang
Intro to FP in Erlang & Message-passing concurrency in Erlang
Message-passing concurrency in Erlang & Sync problems with message-passing
Lock-free programming, parallel linked lists, and parallel queues
Software Transactional Memory
Weak Memory Models

We hold tutorials in Java (emphasis on concurrency) and in Erlang. The last lecture will be used for revision. We are trying to organize a guest lecture.

Labs

Setup
Trainspotting
CCHAT
A-mazed

Examination and grading

We will asset your knowledge of the content of the course both in the labs and in the final written exam. The course awards a total of 100 points: a maximum of 70 points for the exam, and a maximum of 30 points for the labs.

You are allowed to bring the following to the exam:

up to 2 textbooks
up to 4 two-sided sheets of A4 paper with notes (printed or handwritten),
an English dictionary

Exam grading

The exam is graded as follows:

Points

Grade

28–41

3

42–55

4

56–70

5

Points	Grade
28–41	3
42–55	4
56–70	5

Labs grading

Each lab submission receives points when it is marked as passing; the amount of points depends on the lab and on how many of your previous submissions of the same lab were rejected.

	Accepted at attempt #1	Accepted at attempt #2	Accepted at attempt #3
Lab 1	8	6	4
Lab 2	8	6	4
Lab 3	8	6	4

Each lab has a distinction part worth 2 points, available only on the first attempt. To get these 2 distinction points, submit a solution to the whole lab including the distinction part, at the first attempt. This allows the TAs to check the distinction part at the demonstration. Minor errors can be fixed in later attempts. If you do not submit a solution to the distinction part at the first attempt, you forfeit the 2 distinction points.

Overall grading

The overall grade is determined as follows:
- To pass the course, you need these points in each of the exam and labs:
  
  Exam
  
  Lab 1
  
  Lab 2
  
  Lab 3
  
  Minimum points to pass:
  
  28
  
  4
  
  4
  
  4
- If you pass the course, the overall grade is determined based on the total points as follows:
  
  Points in Exam + Labs
  
  Grade
  
  40–59
  
  3
  
  60–79
  
  4
  
  80–100
  
  5

	Exam	Lab 1	Lab 2	Lab 3
Minimum points to pass:	28	4	4	4

Points in Exam + Labs	Grade
40–59	3
60–79	4
80–100	5

Learning objectives and syllabus

Knowledge and understanding:

demonstrate knowledge of the issues and problems that arise in writing correct concurrent programs;
identify the problems of synchronization typical of concurrent programs, such as race conditions and mutual exclusion.

Skills and abilities:

apply common patterns, such as lock, semaphores, and message-passing synchronization for solving concurrent program problems;
apply practical knowledge of the programming constructs and techniques offered by modern concurrent programming languages;
implement solutions using common patterns in modern programming languages

Judgment and approach:

evaluate the correctness, clarity, and efficiency of different solutions to concurrent programming problems;
judge whether a program, a library, or a data structure is safe for usage in a concurrent setting;
pick the right language constructs for solving synchronization and communication problems between computational units.