2015 Papers
Level 4 Papers >> All Papers
COMP453A Extremely Parallel Programming

15 Points

This paper covers advanced parallel programming for large-scale parallelism. A variety of programming techniques will be covered, with application to cluster computers, GPU computing, many-core computing. This paper will be taught using a mixture of classroom and online delivery modes.

Learning Outcomes
Students will be able to explain the basic concepts, benefits, and challenges of parallel programming. They will be able to develop simple parallel programs using a variety of techniques, such as the Hadoop map-reduce framework, the OpenCL language for GPU programming, and Java threadpools. They will understand key Java performance issues and be able to suggest ways of measuring and improving performance.

Pre Requisite Papers
COMP203 Programming with Data Structures or
COMP241 Software Engineering Development
COMP204 Object-Oriented Program Design or
COMP242 Software Engineering Process,
or equivalent Java and JUnit experience

Corresponding Papers
COMP553 Extremely Parallel Programming

Lecturers and Tutors

A Semester

Official Timetable Information

All students should attend the weekly two-hour lab on Thursday, 1-3pm in Lab 1 (R Block).

On weeks 3, 7, 10 and 12, there will be lectures at Thursday 11am-12pm (AG.30) and Friday 9-10am (I1.01), while the other weeks of the semester will be taught online using a flipped classroom approach.

The expected workload is twelve hours per week, for 12.5 weeks = 150 hours.

Required Reading
The required and recommended reading for the paper will be specified on the Moodle website. There is no required textbook for this paper.

Online Support
All online resources, support and discussion forums are available via Moodle.

Means of Assessment
This paper is internally assessed. There is no external exam.

Assessment Ratio
Internal assessment/final examination ratio 1:0

Assessment Deadlines
The assessment schedule is as follows. All submissions are via Moodle, and are due at 11.55pm on the indicated date.

Module Component Percentage Due Date
Java performance
(Week 1)
Photo-Editor Assignment 10 Wed 11 March 2015
(Weeks 2-5)
Map-Reduce Concepts Quiz 5 Wed 18 March 2015
Hadoop Lab Exercise 5 Wed 25 March 2015
Map-Reduce Applications Quiz 5 Wed 1 April 2015
Hadoop Assignment 15 Wed 22 April 2015
GPU Programming
(Weeks 6-8)
GPU Lab Exercise 5 Wed 29 April 2015
GPU Programming Quiz 5 Wed 6 May 2015
GPU Assignment 15 Wed 13 May 2015
Java Threadpools
(Weeks 9-11)
Threadpools Lab Exercise 5 Wed 20 May 2015
Threadpools Quiz 5 Wed 27 May 2015
Threadpools Assignment 15 Wed 3 June 2015
Implicit Parallel Languages (Week 12) Implicit Parallelism Assignment 10 Wed 10 June 2015
TOTAL:   100%  

Attendance Policy
Class attendance is expected. The course notes provided are not comprehensive, additional material will be covered in class. You are responsible for all material covered in class.

Academic Integrity
Follow this link for Academic Integrity information and this link for detailed explanation of How to prevent plagiarism in Computer Science assessment items.

Performance Impairment
Follow this link for information on Performance Impairment.

Student Concerns and Complaints
Follow this link for Student Concerns and Complaints information.

Application for Extension
Follow this link for information on applying for an Extension.

Review of Grade
Follow this link for information on applying for a Review of Grade.

University Regulations
Your attention is drawn to the following regulations and policies, which are published in the University Calendar:

  2007 FCMS. The University of Waikato - Te Whare Wananga o Waikato