NHE2530FNW PCA, Clusters and Grid

1HE UNIVERSITY OF HUDDERSFIELD School of Computing and Engineering

ASSIGNMENT SPECIFICATION Module

DetailsModule CodeNHE2530FNWModule TitlePCA, Clusters and GridsCourse Title/sBEng (Hons) Electronic Engineering and Computer SystemsAssessmentWeighting, Type and Contact DetailsTitlePython for Parallel computing project Weighting26% Mode of working forassessment taskIndividualNote: if the assessment task is to be completed on an individualbasis there should be no collusion or collaboration whilst working onand subsequently submitting this assignment.

Module LeaderDr S AMAMRA [email protected]

odule Tutor/sDr A FARZAMNIAubmissionSubmission and Feedback DetailsHand-out date

16/09/2024How to submit yourworkBrightspace/TurnitiSubmission date/s17/11/2024 by 12:00 noon – if you have any technical issuesubmitting your work, please contact the Module Leader assoon as possible.DO NOT use the work of another student - this includesstudents from previous years and other institutions, as welascurrent students on the moduleDO NOT make your work available or leave insecure, forother students to view or use.

3. Any examples provided by the module tutor should bappropriately referenced, as should examples from externalsources.Further guidance can be found in the SCEN Academic SkillResource and UoH Academic Integrity Resource module inBrightspace.Guidance on using AI:Level – Not PermittedThe use of AI tools is not permitted in any part of thiassessment.School Guidance andSupporIf you experience difficulties with this assessment or with timemanagement, please speak to the module tutor/s, yourPersonal Academic Tutor, or the Student Progress Mentors.Student Progress Mentor – useful links.3Additional GuidancInformatioDetails

• Brightspace Module - SCE Student Progress Mentors(hud.ac.uk).

• Email - [email protected] an appointment - http://hud.ac/rglRequesting a LateSubmissionIt is expected that you complete your assessments by thepublished deadlines. However, it is recognised that there canbe unexpected circumstances which may affect you being ablto do so. In such circumstances, you maysubmit a request foran extension.

Extension applications must be submitted before the publisheassessment deadline has passed.To apply for an extension, you should access the ExtensionSystem on MyHud.ExtenuatingCircumstances (ECs)An EC claim is appropriate in exceptional circumstances, whenan extension is not sufficient due to the nature of the request.

You can access details on the procedure for claiming ECs, on

the Registry website; Consideration of Personal Circumstances- University of Huddersfield, where you can also access theEC Claim Form.You will need to submit independent, verifiable evidence foryour claim to be considered.Once your EC claim has been reviewed you will get an ECoutcome email from Registry.An approved EC will extend the submission date to the next

assessment period (e.g July resit period).Late代 写NHE2530FNW PCA, Clusters and Grid  Submission(No ECs approved)Late submission, up to 5 working days, of the assessmentsubmission deadline, will result in your grade being capped ta maximum of a pass mark.Submission after this period, without an approved extension,will result in a 0% grade for this assessment component.AdditionalGuidanceInformationDetailsTutor ReferralavailabYESResources

• Please note: you can access free Office365 softwareand you have 500 Gb of free storage space available onMicrosoft’s OneDrive – Guidance on downloading Office365.PythoAssignment Title Computer Cluster and Cloud Project

1. Assignment AimsThe aim of this assignment is to investigate parallel computing / processing using Python and it is based on practical laboratory work that consist of three distinct components:

• Test and validation of a serial Python Program to find roots of a QuadraticEquation.

• Parallel a serial program in the above point using Multiprocessing packageon Python.

• Parallel a serial program in the above point using Joblib package on Python.
• Compare Multiprocessing and Joblib performance on Python to parallelserial programs.

2. Learning Outcomes:

Knowledge and Understanding Outcomes

2. Examine and compare the parallel computing packages on Python, and assess theirperformance, and theoretical speed up of parallel processing applications. Ability Outcomes

3. Deploy a parallel processing program, using Python, using environment software.
4. Evaluate different Python packages for parallel processing using appropriatepackage tools, in order to justify their applicability/non-applicability in running specific engineering or scientific application.

3. Assessment Brief

This is an individual assignment, assessed by individual laboratory report. The following serial code (Fig.1) is to find the Roots of a Quadratic Equation. Themathematical representation of a Quadratic Equation is ax²+bx+c = 0. A QuadraticEquation can have two roots, and they depend entirely upon the discriminant.

• If discriminant > 0, then Two Distinct Real Roots exist for this equation
• If discriminant = 0, Two Equal and Real Roots exist.6
• And if discriminant < 0, Two Distinct Complex Roots exist

The serial program Figure.1Assignment objectives:

• Discuss parallel computing on Python with available different libraries?
• Please specify the difference between process and thread.
• Find out the number of your processors on your computer using the multiprocessingpackage.

• Use multiprocessing package to parallel the code in Figure-1 and record the runningime. Hint: You may need to check out the pool.apply function.

• Use joblib package to parallel the code in Figure-1 and record the running time.
• Produce a laboratory report detailing

1. Research on the use of Python for parallel processing.
2. Development of a code using multiprocessing package.
3. Development of a code using joblib package.
4. Performance analysis for the developed programs.
5. Benchmarking – comparison of the developed programs.

Marking Scheme

The assignment work will be assessed through laboratory report.

The laboratory report should have a well-defined structure similar to the following:

Organisation and content (70%)