Dr. Kris Stewart (stewart@rohan.sdsu.edu)
BA 223 594-0491
Office Hours: MW 1-2:30 BA223 & By appointment in LA 073
Return to CS 575 Class Home page
Text: High Performance Computing, Second Edition,
Kevin Dowd & Charles Severance, O'Reilly & Associates, 1998
TOC Our Textbook Table of Contents
Email Repository Email
This course will employ the cooperative learning style to provide students
with experience in how working groups operate in the technical job market
today. Attendance will be taken at each class meeting and will count in
your course grade. Participation in Group Activities will be evaluated
by your peers and will also count in your course grade. We will begin with
two individual computational experiments with lab reports (using instructor
provided template for HTML for reports in your class account). Follow on
activities will be performed in the group setting. Groups will be self-selected
by you to suit your own schedules and backgrounds, stressing
the importance of developing interpersonal skills. Often class time will
be allocated to provide time for lecture (45 minutes) and group discussion
(30 minutes) which is why class attendance is important. There will be
short (10 minute) unannounced quizes in class to ensure students keep up
with reading assignments. The specific sections in our text will be indicated
below because students are not responsible for overly detailed information
which is not critical to this interdisciplinary, introduction to High Performance Computing.
This URL: stewart.sdsu.edu/cs575/calendar_4dec02.html
| Week | Monday | Wednesday | Outcome |
| 1 4sept | Labor Day Holiday | Distribute syllabus and Overview | Acquaint students with goals of course | 2 9sept |
Ch. 1. What is High Performance Computing Ch. 6 Timing (p. 100-107) Lecture covers Chapter 2 Ch2 HPC of text. You should also read the first pages of Chapter 6 on Timing. We discuss Chap. 6 "Profiling" later in the semester |
BA113: Lab 1 Introduction to Xterms and Timer s. This exercise introduces your first computational experiment, which is due via email on 25 Sept 02 | Understanding of timing metrics starts to build. |
| 3 16sept |
More on performance today, (16Sept02 lecture), to connect our text Chapters 2 and 6 with our lab exercise last week. | BA113: Lab 2 First Report and File I/O. | Coding to focus timing results and writing web page with conclusions |
| 4 23sept | Chapter 3,
Ch 03 Memory, in our text starts the discussion of Memory systems for HPC. |
BAM113 Lab 3: Student Survey and Information Overload | Processor speed plus memory speed affect performance. |
| 5 30sept | Ch 04 Floating Point Numbers, gives an
overview of computer representation of values and the importance of floating point numbers and arithmetic. |
BAM113 - Final Lab in BAM 113 (for a while).
Some runtime examples of floating point values, background for the
Second Computational Experiment & Report and tips on
effectively using the compilers, to prepare for
Chapter 5 What a Compiler Does, which we cover in lecture next
week.
Lab 4: Floating Point Hands-on Example and Prep for Using Compilers |
Scientific Programming focussed on floating point computations. |
| 6 7oct |
Ch 05 What a Compiles Does, introduction
to some of the steps a compiler takes in translating your higher-level language to the machine code of the compute platform you use. |
Lecture in BAM256
Grading Rubric, a short
overview of how written reports are scored
Report2_Diffusion background lecture begins the discussion for your next computational experiment. | First graded report is returned |
| 7 14oct | Ch 06 Timing and Profiling. We began the semester examining the first few pages of Chapter 6 to gain background on timing. This week we continue in more depth. Using gprof on the diffusion code, also used for second report. Report2 Assignment, an exploration using Linpack benchmarks applied to the Diffusion Model - Due 30 Oct 02 - Web page and Hard Copy in class. |
Lecture covering Chapter 7: Eliminating Clutter. | Second individual computational experiment assigned |
| 8 21oct | BAM256
Ch 8: Loop Optimizations
Groups begin forming for first group project on Computer Network Security due in class 4Nov02 |
Ch 9: Understanding Parallelism
Ch 09 Highlights
Group Presentation Assignment: Plain-text passwords: benefits, dangers, how to implement Secure Shell (or other secure measures) - DUE 4nov02 Good starting points: security.sdsc.edu security.sdsu.edu | Develop appreciation of Computer Security to protect the community of users |
| 9 28oct |
Frederic Vecoven presents
details on Sun Architectures Frederic Vecoven Sun Microsystems Our SDSU Rohan system is a SunFire 4800 using the FirePlane Bus with 8 GB of RAM, 8 CPUs [distributed on 2 boards]) Ch 10 Shared-Memory Multiprocessor / highlights |
Email Postponement of Report2 to 6 Nov 02 Lab 4 FP Computing and Compilers (introduced the Stommel Model for Ocean Circulation) Second Individual Report Due 30Oct02 / notify instructor when web-page report is ready to be viewed. Bring hard copy to class for another class member to read and critique. |
Begin Fundamentals of Parallel Programming |
| 10 4nov | Groups presentation on Computer Security | BAM113 Lab to select next computational experiment as a group to
run on Rohan Stommel Model and MPI DUE: 27Nov02 |
Start to appreciate complexity of parallel programming |
| 11 11nov |
Ch 11 Programming Shared-Memory Multiprocessors Highlights VI: Scalable Parallel Programming Ch 12 Large-Scale Parallel Processing Ch12 Highlights Amdahl's Law, Interconnect Topologies, MIMD & SIMD Ch 13 Language Support for Performance Ch13 Highlights Data Parallel Problem: Heat Flow |
|
Tools & Examples to Support Parallel Programming |
| 12 18nov |
Guest Lecture (
Jeff Sale): Visualization of the Results of Scientific Programming |
Groups working together on projects | Benefits and Challenges of working with fellow students |
| 13 25nov |
MPI on Rohan,
sample "hello" in C and Fortran codes and their execution Sun Microsystems sample codes, need to update your .login HotPage at NPACI/SDSC MPI: SDSC Workshop Intro to MPI (Message Passing) BH Guide Distribute logins for NPACI Blue Horizon Scalable Parallel System at the San Diego Supercomputer Center. New User Information from NPACI Security Policy and links to safe software. NOTE: Instructor suggests you use ssh and scp from Rohan. You might use the program putty when not on Rohan. NOTE: Ch. 14 of our text uses the red/black ordering to generate examples for Message-Passing Environments for PVM and MPI. Since our class focus will be on MPI and on using the Stommel Model, this material may be not help our focus. |
Group project on Rohan due Group Web Page and Hard copy 27Nov02
| Thankgiving Thur/Fri |
| 14 2dec | Use Rohan to implement 2-D domain decomposition Updated email 28Nov02 Essential POC cc: group members when notifying stewart@rohan.sdsu.edu that final group report is ready to be viewed. Essential Each student email stewart@rohan.sdsu.edu indicating their Individual Contribution to Group Project |
Additional background on tools to assist programmers using NPACI Blue Horizon | Hands-on Experience with a supercomputer |
| 15 9dec | Class Discussions on Progress Port your program to NPACI Blue Horizon and begin looking for performance to compare with Rohan runs |
Groups Finalize Projects We have BAM113 reserved for our class meeting this day.; |
Performance Enhancements Deliverable through Parallel Programming |
| Finals Week | Finals Week | Finals Week | |
| Final Group Project (code and report) due online at Scheduled Final Date | Final: Friday 20 Dec 02 3:30-5:30pm |
Academic Dishonesty will Results in an F in this Course