Course syllabus
Note that the Parallel Functional Programming Course will run remotely in LP4 2021.
Course-PM
DAT280 / DIT261 Parallel functional programming lp4 vt21 (7.5 hp)
The course is offered by the department of Computer Science and Engineering
Contact details
examiner
- Mary Sheeran (mary.sheeran@chalmers.se)
teachers
- John Hughes (rjmh@chalmers.se)
- Mary Sheeran (mary.sheeran@chalmers.se)
TAs
- Maximilian Algehed (algehed@chalmers.se)
- planned office hours from week 2: thursdays 09.00-10.00
- Robert Krook (krookr@chalmers.se)
- planned office hours from week 2: mondays 15.00-16.00
When using office hours, it is a good idea to send a mail, preferably the day before, with a good explanation of your problem and of what you have tried to do to solve it. It may turn out that an email discussion is sufficient to solve your problem. Please do not expect instant turnaround, however. Our PhD students have a lot to do, and need to avoid fragmentation.
Student Representatives
Please volunteer to be a student representative! We will call for volunteers in the first two lectures. Volunteering is particularly important this year, as we are giving and taking the course in difficult circumstances.
The following students have kindly agreed to be student representatives:
Jiahui Le katele1013@outlook.com MPCAS
Emil Logren emil.logren@gmail.com GU
Lukas Skystedt lukassk@student.chalmers.se MPALG
Johan Taube j.taube@student.chalmers.se MPCSN
Please feel free to mail Mary and John, or one of the student representatives if you have questions or concerns about the course.
Mid course meeting
Guest Lecturers
- Troels Henriksen, designer and implementor of Futhark, (Links to an external site.) DIKU Copenhagen
- Gabriele Keller, expert on DSLs for parallel programming, Utrecht University
- Peter Sestoft, Java expert, ITU Copenhagen
- Sven Bodo Scholz, co-developer of Single Assignment C, Raboud University
- Erik Stenman, Erlang expert and entrepreneur
- Richard Carlsson, Erlang expert, Klarna
Course purpose
The aim of the course is to introduce the principles and practice of parallel programming in a functional programming language. By parallel programming, we mean programming using multiple hardware cores or processors in order to gain speed. The course covers approaches to parallel functional programming in both Haskell and Erlang. It covers current research on these topics, and also research on other approaches. The course relies heavily on scientific papers as its source materials and is intended also to give students some insights into research in computer science by considering the development of a sub-field over time. Guest lectures from both academia and industry help to place the presented research and programming methods in context.
Schedule
This course will be given without physical meetings or lectures. We will hold the lectures at the set times using zoom. Please make sure that you know how to use it (though you will be able to attend lectures by just clicking a link).
Here is the schedule in TimeEdit. Ignore the room indications. Everything will be in zoom.
(There will be no teaching on holiday or self-study days, and the schedule will be adjusted accordingly.)
Course literature
This course does not have a set book. Instead, you will be expected to read a number of research papers. Links to those papers will be made available through the lectures page. However, note that Simon Marlow’s book on Parallel and Concurrent Programming in Haskell (Links to an external site.) covers a good chunk of the course (and lots of other interesting stuff). The book is currently free to read online at the O'Reilly site. A student pointed out that you can also buy a pdf for 89 SEK via Google Books.
The lectures appear under Modules and we will also add them to the calendar with the zoom invitations. Click on a lecture and you will get a description of what the lecture is about, related reading material and the slides.
Course design
The course has four compulsory lab exercises, which should be done in pairs. Only in special circumstances will we permit students to work alone, as we must conserve scarce TA resources. Since everyone will be working remotely this year, we will only very rarely give a dispensation for the 2-person lab group rule.
We will use fire for submission and grading of labs. The submission link will be pfp-lp4-21.fire.cse.chalmers.se. The lab exercises are central to your learning on this course. Those who take them seriously typically also pass the written exam easily. That is our intention.
You must pass all four lab exercises (as well as the final written exam) to complete the course.
The course has two main parts, each of which in turn has two parts, and hence the four lab exercises.
- Parallel Programming in Haskell
- Robust Parallel Erlang Programming
- Data Parallel Programming
- Map Reduce and noSQL databases
The guest lectures on the course are not just entertaining additions but contain examinable material. For example, there is a lab exercise related to the lecture on Futhark. So come to the guest lectures! They also broaden the view of parallel functional programming presented by the course.
Changes made since the last occasion
The course will be very similar to last year's instance, including how the exam is conducted.
Learning objectives and syllabus
Learning objectives:
Knowledge and understanding
1. Distinguish between concurrency and parallelism.
2. Give an overview of approaches to parallelism in functional programming languages in the scientific literature.
Skills and abilities
1. Write, modify and test parallel functional programs, to run on a variety of architectures such as shared memory multiprocessors, networks of commodity servers, and GPUs.
2. Interpret parallelism profiles and address bottlenecks.
Judgement and approach
1. Identify when using a functional language may be appropriate for solving a parallel programming problem.
2. Select an appropriate form of parallel functional programming for a given problem, and explain the choice.
Link to the syllabus on Studieportalen.
TODO The study plan at GU should be linked here.
Examination form
You must complete all four lab exercises to complete the course. Labs are done in pairs.
You must also pass the written exam at the end of the course. Your grade on the course is determined entirely by your result on the written exam. The written exam will have to be taken online, and this has resulted in greater emphasis in being able to write, run and measure parallel programs than in previous years.
Schedule
Course summary:
| Date | Details | Due |
|---|---|---|