Course syllabus
Course-PM
DAT630 Project in parallel computing lp3 VT25 (7.5 hp)
Course is offered by the department of Computer Science and Engineering
Contact details
- Miquel Pericàs <miquelp@chalmers.se> (Examiner + Lecturer)
- Hari Abram <hariv@chalmers.se> (Teaching Assistant)
- Sonia Rani Gupta <soniar@chalmers.se> (Teaching Assistant)
Course purpose
This course looks at the design of high-performance and energy efficient software for parallel computers, performance modeling, and HW/SW codesign of parallel and heterogeneous computer systems. The course is a project-oriented course in which you will develop skills for programming and assessing the performance of parallel and heterogeneous computers. A series of lectures on the principles of parallel hardware provide required background for architecture-aware program optimization. Via group assignments you will furthermore develop your teamwork skills, including version control, communication and presentation.
Schedule
Check TimeEdit for slots and rooms. Wednesday 8h sessions are self-study session. Lab rooms are allocated which you can use to work on your projects. In addition. we come 9h-10h every week allocated to project supervision. You can meet us in our offices (or zoom) to discuss about your progress.
The following schedule is likely to change! Check periodically. In general, we will try to notify changes well ahead of the affected sessions.
| When | Where | What |
Responsible |
|
| #1 | Jan 21st, 10-12h | ES51 | Introduction to DAT630 and Project Description | Miquel |
| #2 | Jan 24th, 9-12h | SB-M415 | Tutorial #1: GIT, Gitlab, CI/CD | Hari / Sonia |
| #3 | Jan 28th, 9-12h | SB-M415 | Lecture #1: Benchmarking, Vector Architectures | Miquel |
| #4 | Jan 31st, 9-12h | EL42 | Supervision | Hari / Sonia / Miquel |
| #5 | Feb 4th, 9-12h | SB-M022 | Lecture #2: Chip Multiprocessors | Miquel |
| #6 | Feb 7th, 9-12h | EL43 | Supervision | Hari / Sonia / Miquel |
| #7 | Feb 11th, 9-12h | SB-M022 | Tutorial #2: Containers, Job Scheduling | Hari / Sonia |
| #8 | Feb 14th, 10h-12h | EF | Supervision | Hari / Sonia / Miquel |
| #9 | Feb 18th, 9h-12h | SB-M022 | Lecture #3: GPGPU | Miquel |
| #10 | Feb 21 | -- | NO MEETINGS ON THIS DAY | |
| #11 | Feb 25, 9h-12h | SB-M415 | cancelled | |
| #12 | Feb 28, 9h-12h | online | Lecture #4: HW-support for MPI | Miquel |
| #13 | March 4th, 9h-12h | SB-M415 | Lecture #5: Hands-on on HW/SW codesign | Sonia |
| #14 | March 7th, 9h-12h | Supervision | Hari / Sonia / Miquel | |
| #15 | March 11th, 9h-12h | no lecture (work on report and presentation) | ||
| #16 | March 14th, 8h-12h | EE | Project Presentations (all teams) | Miquel |
Course design
- version control and testing
- containers and singularity
- advanced performance analysis and optimization
- tools for HW/SW codesign
- multicore, manycore and vector architecture, with a focus on the memory hierarchy and its implications to the performance of OpenMP and vector programs
- architectural support for message passing, and how it impacts the performance of MPI programs
- architecture of GPGPUs, and how it impacts optimization of CUDA programs
Continuous feedback will be collected to assess the degree of learning and to improve future versions of the course.
The course will use Canvas as the main tool for communication. Chalmers Gitlab will be used for software development.
Changes made since the last occasion
This is the first edition of this course
Learning objectives and syllabus
- describe current approaches to parallel computing
- explain the design principles of the hardware support for the shared memory and message passing programming models
- describe the implementation of different models of thread-level parallelism, such as chip multiprocessors, many-cores or GPGPUs
- list basic approaches to software engineering, performance engineering and teamwork
- design scalable parallel software and analyze its performance
- debug and optimize the execution of parallel programs on heterogeneous computers
- apply performance modeling and simulation techniques for performance prediction and HW/SW codesign.
- apply teamwork skills for version control, communication, presentation, and conflict management.
- ability to cooperate in diverse group compositions with team members with different skills, cultural and educational backgrounds, gender and nationality
- analyze the trade-offs of different approaches to parallel computing in terms of function, performance and cost
- analyze and judge the performance portability of a piece of software
Link to the syllabus on Studieportalen.
https://www.chalmers.se/en/education/your-studies/find-course-and-programme-syllabi/course-syllabus/DAT630/?acYear=2025/2026
Examination form
The course has two components: written examination (1.5c) and project (6c)
The written examination will consist of a 2h exam covering lectures 1-6. The written exam has only two grades: pass/fail. Bonus points will be given throughout the course that impact the written exam.
The project will be evaluated based on multiple criteria:
- quality of the porting and optimization effort (1-5)
- application of versioning, containers and performance analysis methodologies (1-5)
- quality of the written report (1-5)
- quality of the presentation (1-5)
- quality of the peer review (1-5)
- individual feedback by the each team's supervisor (1-5)
A grade of 3 needs to be achieved in each of the four categories in order to get a project grade. The final grade will be the average of the four grades, with rounding to the nearest full grade. Grades 3.5 and 4.5 will be rounded upwards.
Both the written examination and project need to be approved in order to get a course grade. The final course grade will be the same as the project grade.
Course summary:
| Date | Details | Due |
|---|---|---|