Read-Me-First (and Syllabus)
|
EE 4683/5683 is an online course in Spring
2023. The course administration will be different for students
registered for this online course. In what follows, course
implementation will be explained, so please read carefully.
Course
Syllabus (PDF)
Course Information
Credits: 3
Days: Mondays
Time: 2:00 PM - 2:55 PM
Location: ARC 321
Webpage: http://oucsace.cs.ohio.edu/~avinashk/classes/ee468/ee468.htm
Course Instructor
Name: Avinash Karanth
Room: STKR 330/335
Phone: (740)-597-1481
E-mail: karanth@ohio.edu
Office Hours: 10:00 AM - 12:00 PM, Monday, or by
appointment.
Grading Assistant: Siqin Liu
Office: STKR 305
E-mail: ls847719@ohio.edu
Textbook
- (Required) Computer Architecture - A Quantitative
Approach, John L. Hennessy and David A. Patterson, 5th Edition.
- (Optional) Computer Organization and Design - The
Hardware/Software Interface, David A. Patterson and John L.
Hennessy, 3rd Edition.
Prerequisites
- EE 3613: Computer Organization
Class Policies
- Class Meeting: As this is an online class,
there will be no regular meeting as all instructional
screencasts, lecture notes and assignments will be posted
online. However, for the convenience of the students (undergrads
and grads), I will periodically meet with the students, every
week on Mondays from 2-3 PM to discuss the critical issues with
the class.
- Lectures: There are a total of 7 modules
with multiple sub-parts covering computer architecture topic.
Each module will be composed of lectures slides (PDFs) and
instructional screencasts (voice-over powerpoint slides). These
will be posted on a weekly basis and students will be notified
via email. All lecture slides will be posted on the course
website; the instructional screencasts will be available via
blackboard. Each screencast will be 10 - 20 minutes and will be
accompanied with powerpoint slides. Since this is an online
course, it is up to you to keep up with the lessons, readings
and assignments. This is critical since much of the material
builds on itself and if you get a few weeks behind, it can be
extremely difficult to catch up.
- Assignment Submission: All assignment
submissions are due according to the due dates indicated in the
course schedule. Periodically, I will send reminders when they
become due. All assignments will be collected via blackboard.
- Exam Submission: For all students
registered for the class, there will be in-person
one-hour midterm exam and two-hour final.
- Make-Up Exam: Make-up exams will only be
given to those with valid university absences such as documented
illness. In such cases, the student is requested to contact the
instructor as soon as possible to schedule a make-up exam.
- Academic Misconduct: Any academic
dishonesty will not be tolerated. Unless otherwise specifically
stated by your instructor, all course work should be done on
your own. Please refer to the OU Student Code of Conduct.
- Regrades: All requests for regrades must be
submitted in writing within one week of the distribution of the
graded material.
- Late Homeworks: Late homework assignment
will be accepted for a maximum of three days after the due date.
For each day your assignment is late, 10% of the total possible
points will be deducted from your score. Typically, homework
assignments are due on Monday, and they will be graded with
reduced points until the following Friday.
Undergraduate/Graduate Grading Policy
- Homeworks (4): 20%
- Midterm
(1): 25%
- Final Exam
: 30%
- Projects (2) : 25%
- Assignments are due via e-mail or through
Blackboard on the date and time specified.
- All grades will be posted on Blackboard.
- All grading is based on the 12-point system.
[100-93] A, [92-90] A-, [89-87] B+, [86-83] B, [82-80] B-,
[79-77] C+, [76-73] C, [72-70] C-, [69-67] D+, [66-63] D,
[62-60] D-, [59-..] F. Instructor reserves the right to lower
the limits above, but I promise not to raise them.
- All exams will be in-class (midterm & final).
Course Outline
EE 4683/5683 is intended to provide undergraduate and
graduate students with an in-depth study of digital systems and
computer design. It provides a basic knowledge and ability
required for understanding and designing computer systems. The
important topics covered in this class include cache and main
memory concepts, virtual memory, multi-level caches, I/O devices,
multiprocessor networks, snoopy and directory cache coherence.
Graduate students are expected to complete a term-paper. The
topics covered in this course are as follows:
- Performance Trade-offs for Multicores
- Advanced Pipelining: Superscalar and VLIW
- Instruction Level Parallelism: Static and Dynamic
Scheduling
- Limits to ILP: Multithreading and Multicores
- Memory Hierarchy and Multi-Level Caching
- I/O Fundamentals and Techniques
- Shared Memory Multiprocessing
- Interconnection Networks & Clusters
Tools
- WinMIPS64 - MIPS64 instruction set simulator for
Windows
- Dinero - Cache simulator runs on Unix, linux or
Cygwin
Tentative Dates
- Midterm Exam: Monday, March 6, 2023
- Final Exam: Monday, May 1, 2023