EECS 224 High-performance Computing (2017-2018)

EECS 224 High-performance Computing

(Not required for any major.)
Catalog Data:

EECS 224 High-performance Computing (Credit Units: 4) Fundamentals of high-performance computing, covering both theory and practice. Topics include performance analysis and tuning, design of parallel and I/O efficient algorithms, basics of parallel machine architectures, and current/emerging programming models (shared memory, distributed memory, and accelerators). Prerequisite: EECS 215 or CS 260. Graduate Students only. (Design units: 0)

Required Textbook:
. Edition, , 1969, ISBN-13 978-1439811924.

. Edition, , 1969, ISBN-13 978-0201648652.

Recommended Textbook:
None
References:
None
Coordinator:
Aparna Chandramowlishwara
Relationship to Student Outcomes
No student outcomes specified.
Course Learning Outcomes. Students will:

1. Students will understand and program parallel computers to solve real-world challenging problems in science and engineering.

2. Students will emerge knowing how to map and understand the behavior algorithms and applications on parallel systems.

Prerequisites by Topic

Knowledge of algorithms and basic algorithm analysis. Knowledge of computer architecture. Students are also expected to have sequential programming experience; and organization of digital computers (EECS 112) or computer systems architecture (CompSci 152).

Lecture Topics:

Performance analysis and tuning, design or parallel and I/O efficient algorithms, basics of parallel machine architectures, state-of-the-art techniques in performance engineering, and parallel programming models for shared memory, distributed memory and accelerators.

Class Schedule:

Meets for 3 hours of lecture and 1 hour of laboratory each week for 10 weeks.

Computer Usage:

Campus-wide HPC cluster.

Laboratory Projects:

Basic tools for building, debugging, and profiling software, OpenMP for multicore processors, MPI for distributed memory machines, collective communication, CUDA for GPUs, and single-node optimization and tuning.

Professional Component
Design Content Description
Approach:
Lectures:
Laboratory Portion:
Grading Criteria:

Homework Assignments 40% Midterm 20% Final Project/Final Exam 40%

Estimated ABET Category Content:

Mathematics and Basic Science: 0.0 credit units

Computing: 0.0 credit units

Engineering Topics: 0.0 credit units

Engineering Science: 0.0 credit units

Engineering Design: 0.0 credit units

Prepared:
June 22, 2017
Senate Approved:
November 9, 2016
Approved Effective:
2017 Spring Qtr