Module Details
The information contained in this module specification was correct at the time of publication but may be subject to change, either during the session because of unforeseen circumstances, or following review of the module at the end of the session. Queries about the module should be directed to the member of staff with responsibility for the module.
Title High Performance Computing
Code COMP328
Coordinator Dr MAS Abo-Tabik
Computer Science
M.A.Abo-Tabik@liverpool.ac.uk
Year CATS Level Semester CATS Value
Session 2021-22 Level 6 FHEQ Second Semester 15

Aims

1. For students to understand the motivation and opportunities of high performance computing, and to have sufficient understanding of topics in order to use HPC facilities;

2. For students to appreciate challenges of obtaining peak performance and how to tackle such challenges;

Learning Outcomes

(LO1) to gain an appreciation of the needs for parallel computing and High Performance Computing (HPC)

(LO2) to be able to read & understand parallel programmes written by others

(LO3) to be able to design and implement parallel programmes, using a variety of paradigms, and to run them on a real world HPC facility

Syllabus
 

The course comprises a mix of lectures and labs during the 12 week teaching. There will be 23 hours of lectures:
• Motivation for HPC, example use cases from industry etc) – 2 lectures
• The use of compiler flags (and limitations) of serial node optimisation – 2 lectures
• Theory of parallelism: opportunities, types of parallelism, barriers to performance and scoping max performance – 4 lectures
• Using parallelised libraries/pre-written applications – 1 lecture
• Using OpenMP for shared memory – 4 lectures
• Introduction to MPI for distributed memory (multi-node) programming – 4 lectures
• Overview of GPU and emerging architectures – 3 lectures
• Work through of assessments – 2 lectures
• Summary/revision – 1 lecture
There will be 18 hours of labs, designed to re-enforce topics from lessons and cover elements useful for the assigned coursework. Labs will make use of the University HPC facilities to give students real world experience. Each lab will be 2 hours long, with lecturer + support, to help maximise learning for all students.

Teaching and Learning Strategies

lectures, covering theory and live demo of codes implementing theiry, with interactive assessment periodically to ensure uptake / help focus where more teaching required

hands on labs to back up lectures and to lead in to assessment work

course books and online references will be expected to be read as part of private study time

Due to Covid-19, in 2021/22, one or more of the following delivery methods will be implemented based on the current local conditions.
(a) Hybrid delivery
Teaching Method 1 - Lecture
Description: Mix of on-campus/on-line synchronous/asynchronous sessions
Teaching Method 2 - Laboratory Work
Description: Mix of on-campus/on-line synchronous/asynchronous sessions

(b) Fully online delivery and assessment
Teaching Method 1 - Lecture
Description: On-line synchronous/asynchronous lectures
Teaching Method 2 - Laboratory Work
Description: On-line synchronous/asynchronous sessions

(c) Standard on-campus delivery
Teaching Method 1 - Lecture
Description: Mix of on-campus/on-line synchronous/asynchronous sessions
Teaching Method 2 - Laboratory Work
Description: On-campus synchronous sessions

Teaching Schedule
  Lectures Seminars Tutorials Lab Practicals Fieldwork Placement Other TOTAL
Study Hours 23

     18

     41
Timetable (if known)              
Private Study 109
TOTAL HOURS 150

Assessment
EXAM Duration Timing
(Semester)		 % of
final
mark		 Resit/resubmission
opportunity		 Penalty for late
submission		 Notes
(328) written exam that will test on each learning outcome and cover all key points of the syllabus.  2 hours    80       
CONTINUOUS Duration Timing
(Semester)		 % of
final
mark		 Resit/resubmission
opportunity		 Penalty for late
submission		 Notes
(328.1) hands-on assessment to write an efficient parallel programme to solve a problem that will be defined and explained to the studies, by making use of the university HPC cluster      20       

Recommended Texts
Reading lists are managed at readinglists.liverpool.ac.uk. Click here to access the reading lists for this module.