Semester

Semester 1, 2019

• Nasser Giacaman
• Oliver Sinnen (coordinator)

Advanced parallel and high performance computing concepts and techniques such as memory architecture and networks; multicores, hardware acceleration devices; shared memory and data parallel programming; object oriented and low level parallel programming; parallelisation process: subtask decomposition, dependence analysis and scheduling. Core concepts are extended by a hands-on research project in which a challenging parallel computing problem is analysed and solved.
Prerequisite: 15 points from SOFTENG 306, COMPSYS 302, MECHENG 313, or 30 points at Stage III in Computer Science

Further notes

The course is divided into two parts:

The first part consists of lectures and tutorials given by the lecturer(s), covering the fundamentals of parallel computing, technologies and practical methods. Test 1 will assess the learning of this first part.

Equipped with the knowledge of the first part, student form small groups and each is given an assignment for the second part of the course during which they work largely independent on that topic. There are numerous well selected topics and each group needs to give a presentation on that topic to the class. This induces peer-learning and exposes the class to more advances topics of parallel computing. Test 2 will cover the material presented during the second part of the course.

Assessment components are:
===
* Test 1, 20%
* Test 2, 20%
* Assignment 60% (consisting of presentation (20%), implementation (20%) and report (20%)).
- the presentation mark (20%) is composed of three components: lecturer mark (8%), peer mark (8%), peer marking participation (4%)

Test material
---
Test 1 covers the material delivered until test 1 and test 2 covers the material delivered after the cut-off for test 1 until test 2. This explicitly includes student presentations during that time.

Assignment groups and topic
---
For the assignment, students need to form groups of three and select one of the assignment topics announced by the lecturers in week 5 or 6. While working in a group, the marking will be individual and all students need to contribute to all components, i.e. to the presentation, implementation and report.

Presentation
---
At the time of the topic allocation, each group will be given a slot in the lectures during which they need to make their presentation. The allocation of topics and slots will be on a first-come-first-serve basis after the topics and slots are announced. Test 2 will be mostly based on the student presentations, hence students need to attend each others presentations.

The presentations will be marked by the lecturer (8%) and also by the students (8%). For the participation in the peer marking, students receive 4% if they have marked enough peer presentations (details to be advised later).

File Repository
---
Each group will use a private file repository (github) for all code, scripts, documents (including report and presentation slides), figures, images and other digital artefacts related to the project. This repository also serves for the submission of the implementation at the due date. Each member should commit their own work during the semester. Indicate joint work in the commit messages. In general, commit messages must be meaningful and will be considered in the marking of the implementation and report. It is NOT acceptable to use another repository during the semester, only making a final commit at the end.

Report
---
One per group, submitted on Canvas. Format: 8 pages, single column, 2.5cm margins, 11 point font. Include table of contributions by each member.

Peer evaluation/feedback
---
A Teammates based peer evaluation will be carried out at the end of the assignment. This will give the students the opportunity to convey feedback to their group members and To the lecturing team. It will also help to more fairly mark the individual contributions of each team members. Students will receive an email towards the end of the semester, inviting them carry out the peer evaluation. This must be completed by all students. A corresponding Teammates invitation email will be send out towards the end of the semester.

Late penalties (assignment implementation and report)
---
24h: -15% of achieved marks
48h: -25% of achieved marks
>48h: 0 marks

No exam.

Students are urged to discuss privately any impairment-related requirements face-to-face and/or in written form with the course convenor/lecturer and/or tutor.

No description given

The University of Auckland will not tolerate cheating, or assisting others to cheat, and views cheating in coursework as a serious academic offence. The work that a student submits for grading must be the student's own work, reflecting his or her learning. Where work from other sources is used, it must be properly acknowledged and referenced. This requirement also applies to sources on the world-wide web. A student's assessed work may be reviewed against electronic source material using computerised detection mechanisms. Upon reasonable request, students may be required to provide an electronic version of their work for computerised review.

All students enrolled at the University of Auckland are required to complete a compulsory Academic Integrity course, usually in their first semester/year of enrolment. The University of Auckland’s full guidelines on procedures and penalties for academic dishonesty are available here.