The University of Western Ontario
London, Canada

Department of Computer Science

CS4435/CS9624b -- High Performance Computing: From Models of Computation to Applications

Course Outline -- Winter 2011

Course Description

Today's parallel hardware architectures (multicores, graphics processing units, etc.) and computer memory hierarchies (from processor registers to hard disks via successive cache memories) enforce revisiting many fundamental algorithms which were often designed with algebraic complexity as the main complexity measure and with sequential running time as the main performance counter.

Indeed, on modern computers minimizing the number of arithmetic operations is no longer the primary factor affecting performance. Effective utilization of the underlying architecture (increasing the amount of instruction level parallelism, better utilizing the memory hierarchy, etc.) can be much more important. Many examples exist where algorithms that use more arithmetic operations outperform algorithms with fewer arithmetic operations.

In this course, we will present different theoretical tools (cache complexity, multithreaded parallelism, space-time tradeoffs, code optimization for parallelism and locality, etc.) that are adapted for taking best advantage of parallel architectures and hierarchical memories.

We will discuss a large variety of applications (finite-difference methods, compression algorithms, artificial intelligence, metabolic networks, etc.) and compare the implementation techniques of a given algorithm on different hardware architectures (multicores, GPGPUs).

To have a more detailed idea about this course, please visit the web site of a course I taught last year and from which I will borrow part of the materials for this year's course: CS 4435 and CS 9624 - High Performance Computing with a Focus on Hardware Acceleration Technologies.

Follow this link for various resources (software tools and tutorials, hardware documentation, conferences, other HPC course web sites, etc.) regarding this course and HPC in general.

Prerequisites for undergraduate students


Name:Marc Moreno Maza
Office Hours: Tuesday 2:30-4:40 pm
Phone:661-2111 x6891

Lecture Notes and Textbook

Notes of each lecture will be available on the course website, approximatively one or two days after the oral presentation. There is no textbook.

Course Website

The course web site is accessible from:

Please check the site often for updates on lecture notes and errata. Also be aware that the course website is not a substitute for actual classroom attendance!

Lecture Topics

The list of topics will be something on the order of:

  1. Analysis of multithreaded algorithms
  2. Cache complexity, I/O complexity
  3. Space-time tradeoffs
  4. Optimizing code for data locality and parallelism
  5. Optimizing stencil computation
  6. Parallel string algorithms (longest common subsequence, string matching) targeting multicores
  7. Parallel graph algorithms (maximum independent set, coloring, transversal) targeting multicores and applications
  8. Implementing FFTs: from multicores to GPGPUs

Class Schedule

Lectures: 3 hours (Monday from 1:30 to 4:30pm in MC316). In general, a lecture consists of a talk by the instructor followed by an exercise session or a quiz.

Each student is expected to attend the lectures. In particular, quizzes (short written tests) may take place without notice.

Student Evaluation

Assignment/Project/Quiz Schedule

All dates are tentative and currently subject to change, although it is doubtful by any significant amount.

Evaluation Technique Weight Posted Date (tentative!) Due Date (tentative!) Workload
Assignment One20%Mo, Jan. 31Mo, Feb. 14light
Assignment Two20%Mo, Feb. 14Mo, Mar. 7light
CS 4435 Project 40%Mo, Mar. 7Mo, Apr. 4regular
CS 9624 Project 1 40%Mo, Jan. 31Mo, Mar. 7regular
CS 9624 Project 2 40%Mo, Mar. 7Mo, Apr. 4regular
Quizzes10% eachN/AvariousN/A

If for any reason the schedule given above cannot be adhered to, the assignment, project and quiz marks will be pro-rated. For instance, if an assignment has to be canceled for any reason, the remaining assignment weight will be prorated to add up to 30%.)

Every effort will be made to have assignments, projects and quizzes marked and handed back within 3 weeks of the hand-in date, preferably sooner.


Quizzes may be held without being announced in advance.

Quizzes will be closed book.


Assignments will be due on the (tentative) dates listed above. The assignment will be sent by email to the instructor.

Extensions will be granted only by the course instructor. If you have serious medical or compassionate grounds for an extension, you should take supporting documentation to the office of the Dean of your faculty, who will contact the instructor.


CS 4435 Projects and CS 9624 Projects 2 will be presented by the students during the class on the (tentative) date listed above.

It is expected that presentation session may last longer than a usual class.

Computing Facilities

Each student will be given an account on the Computer Science Department senior undergraduate computing facility, GAUL. In accepting the GAUL account, a student agrees to abide by the department's Rules of Ethical Conduct

Note: After-hours access to certain Computer Science lab rooms is by student card. If a student card is lost, a replacement card will no longer open these lab rooms, and the student must bring the new card to a member of the Systems Group in Middlesex College Room 346.

Email Contact

The instructor will occasionally need to send email messages to the whole class, or to students individually. Email will be sent to your GAUL email address. You must make sure that you read your email on GAUL on a frequent and regular basis, or have it forwarded to an alternative email address if you prefer to read it there.

However, you should note that email at ITS (your UWO account) and other email providers such as or may have quotas or limits on the amount of space they can use. If you let your email accumulate there, your mailbox may fill up and you may lose important email from the instructor. Losing email that you have forwarded to an alternative email address is not an excuse for not knowing about the information that was sent.

Students can ask questions via email, however if there are any large, somewhat complicated issues, it is recommended to discuss them during office hours. Moreover, you MUST use your UWO account or your GAUL account in order to write to the instructor. (Emails from non-academic accounts will be automatically ignored.) In addition, you must include "CS4402" in the subject line.

Academic Accommodation for Medical Illness

If you are unable to meet a course requirement due to illness or other serious circumstances, you must provide valid medical or other supporting documentation to your Dean's office as soon as possible and contact your instructor immediately. It is the student's responsibility to make alternative arrangements with their instructor once the accommodation has been approved and the instructor has been informed. In the event of a missed final exam, a "Recommendation of Special Examination" form must be obtained from the Dean's Office immediately. For further information please see

A student requiring academic accommodation due to illness should use the Student Medical Certificate when visiting an off-campus medical facility or request a Record's Release Form (located in the Dean's Office) for visits to Student Health Services. The form can be found at

Accessibility Statement

Please contact the course instructor if you require material in an alternate format or if you require any other arrangements to make this course more accessible to you. You may also wish to contact Services for Students with Disabilities (SSD) at 661-2111 x 82147 for any specific question regarding an accommodation.

Ethical Conduct

All assignments are individual assignments. You may discuss approaches to problems among yourselves; however, the actual details of the work (assignment coding, answers to concept questions, etc.) must be an individual effort. Assignments that are judged to be the result of academic dishonesty will, for the student's first offense, be given a mark of zero with an additional penalty equal to the weight of the assignment also being applied.

Students must write their essays and assignments in their own words. Whenever students take an idea, or a passage from another author, they must acknowledge their debt both by using quotation marks where appropriate and by proper referencing such as footnotes or citations. Plagiarism is a major academic offense (see Scholastic Offense Policy in the Western Academic Calendar).

The University of Western Ontario uses software for plagiarism checking. Students may be required to submit their written work and programs in electronic form for plagiarism checking.

Scholastic offenses are taken seriously  and students are directed to read the appropriate policy, specifically, the definition of what constitutes a Scholastic Offense, at the following Web site:

Marc Moreno Maza
Last modified: Sun Jan 3 0:33:44 EDT 2009