Applied Mathematics 483/583

High Performance Scientific Computing

Spring Quarter, 2014

Coursera Version: AMath 483 and 583A are currently full and we cannot add more students due to room size limitations. An alternative if you want to learn the material without receiving credit is to register for the Coursera version of this class, which was first given in 2013 and is repeating starting on March 31, 2014. This version uses the same lecture videos from 2013 that will be used used in 483/583.

Quick links:

Canvas page for registered students (videos of lectures, office hours, discussion board, quizzes, scores, etc.)
Class notes
slides
Lab 1 Please read before lab on Tuesday April 1.
uwhpsc.zip (803 MB) Virtual machine (see notes on using the VM.

Class Info:

Professor Randy LeVeque will be "flipping the class" this year.
Students will be required to watch 3 hours of lectures a week that will be available on the web (videos recorded in Spring, 2013). These follow the slides and lecture notes from last year, but some new material will also be developed.
For on-campus students, the class meets T-Th, 2:30 - 3:20 in OUG 136 for lab sessions. This is an Active Learning Classroom and class time will be used primarily for working together with other students on exercises designed to illustrate and reinforce the material. Some additional new material will also be presented. Attendance will be required at these sessions.
Undergraduates should enroll in AMath 483 and graduate students in 583. The lectures and classroom sessions are identical, but the 583 course will have some additional and/or more advanced assignments. The class sizes are capped due to the size of the room and additional students cannot be accomodated.
Section 583B is for online Masters degree students only and is not available to on-campus students. Students in this section will view the same video lectures as on-campus students. In addition, parts of the T-Th classroom sessions will be taped and available to watch, along with the exercises tackled by students in these sessions.
Homework and a final project will consist primarily of programming assignments. There will be quizzes on the lectures and on-campus students will often submit something during lab sessions, but there will be no major exams.
No textbook is required. Class notes are available online.
All software to be used is freely available and open source. All of the computing required for this class can be done freely in the cloud using SageMathCloud. A demo will be given during the first lab session.

Description

Introduction to hardware, software, and programming for large-scale scientific computing. Overview of multicore, cluster, and supercomputer architectures; procedure and object oriented languages; parallel computing paradigms and languages; graphics and visualization of large data sets; validation and verification; and scientific software development.

[More about the class and syllabus]

Recommended Background

Experience writing and debugging computer programs is required --- preferably experience with scientific, mathematical, or statistical computing, for example in Matlab or R. (Previous knowledge of Fortran, Python, or parallel computing languages is not assumed.)

Students should also be comfortable with undergraduate mathematics, particularly calculus and linear algebra, which is pervasive in scientific computing applications. Many of the examples used in lectures and assignments will require this background. Past exposure to numerical analysis is a plus.