Syllabus for Introduction to Parallel Programming

Learning outcomes

On completion of the course, the student should be able to:

• explain and recognize basic concepts of concurrent and parallel computation,
• write and modify programs that use threads and locks to exploit shared-memory concurrency,
• explain common issues in shared-memory programming and use synchronization to avoid these issues,
• implement solutions for fundamental synchronization problems,
• explain consistency models and apply them to reason about the correctness of concurrent programs,
• identify potential for concurrency or parallelism in sequential algorithms, and reimplement them accordingly.

Content

Basic concepts of concurrent and parallel computation, e.g., speedup and parallel slowdown, Amdahl's law, Flynn's taxonomy, instruction-level parallelism, data independence. Shared-memory concurrency based on threads and locks, and its use in a language such as C++. Associated concepts and issues, e.g., synchronization, non-determinism, data races, deadlocks, fairness. Fundamental synchronization problems, such as the producer-consumer problem, and techniques for their solution. Consistency models such as linearizability and sequential consistency. The fork-join model. Task parallelism vs. data parallelism.

Instruction

Lectures, labs.

Assessment

Oral and written examination of assignments. Written exam.

If there are special reasons for doing so, an examiner may make an exception from the method of assessment indicated and allow a student to be assessed by another method. An example of special reasons might be a certificate regarding special pedagogical support from the disability coordinator of the university.

Other directives

The course cannot be included in the same degree as 1DL540 Language Abstractions For Concurrent and Parallel Programming, 10 ECTS credits, or 1DT096 Operating Systems and Process Oriented Programming.