Aktualności / News > Dydaktyka / Teaching > Historia / History > Parallel Programming, summer 2017/18
Parallel Programming, summer 2017/18
Lecture
Planned schedule of lectures:- 23.02 basic notions in parallel processing (the change in the schedule of lectures in not included in the slides!)
- 27.03 classical problems in synchronization, semaphores
- 06.03 semaphores, continued
- 13.03 monitors
- 20.03 parallell computational models, parallel algorithms
- 27.03 parallel algorithms, continued
03.04 spring holidays - 10.04 models of distributed computation, distributed algorithms
- 24.04 final test
During the final test you can obtain up to 50 point, at least 25 are needed to pass the course. The final grade is computed from the points from the test and from the labs.
Labs
During the labs the student have to solve four tasks, each during 3-4 weeks.
Submission of each task consists of two parts:
- submitting the files with the solution before the deadline. To submit the files, please use the uploader. The files should be zipped, the filename should be pp_[your_login]_tasl[task-no].
The solution should be submitted before 7 AM on the day of the deadline. - presenting your solution during the classes, discussion, answering questions.
Submitting the files and not presenting the solution is equivalent to failing the task. Tasks 1 and 2 are worth 10 point each, tasks 3 and 4 are worth 15 points each. Each (started) week of being late results in -5 pt penalty.
To pass the course, you need to get at least 25 points from labs. It is not necessary to submit all tasks.
The classes that are not associated with deadlines are consultation. If I am not in the classroom, look for me in my office (515).
26.02-02.03 - task 1 is givenPlanned schedule of labs:
19.03 - deadline for task 1, task 2 is given
09.04 - deadline for task 2, task 3 is given
07.05 - deadline for task 3, task 4 is given
04.06 - deadline for task 4
11.06 - the last deadline for anything
Tasks for labs
- task 1: semaphores
- comment: the production process, once started, cannot be terminated; curses only affect starting the production
- task 2: monitors
- task 3: distributed algorithms
- task 4: map-reduce