0% found this document useful (0 votes)
97 views

ECE690 Syllabus

This document is a course syllabus for ECE690: Computer Systems Architecture at NJIT. The course will cover advanced microprocessor and computer system design topics like pipelining, instruction-level parallelism, data-level parallelism, and thread-level parallelism. Students will learn how to evaluate hardware performance and estimate power dissipation. The textbook is Computer Architecture: A Quantitative Approach by Hennessy and Patterson. Grading will be based on a midterm exam, final exam, and design project report. Students will gain an understanding of modern parallel computer architectures.

Uploaded by

yeswanth1234
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
97 views

ECE690 Syllabus

This document is a course syllabus for ECE690: Computer Systems Architecture at NJIT. The course will cover advanced microprocessor and computer system design topics like pipelining, instruction-level parallelism, data-level parallelism, and thread-level parallelism. Students will learn how to evaluate hardware performance and estimate power dissipation. The textbook is Computer Architecture: A Quantitative Approach by Hennessy and Patterson. Grading will be based on a midterm exam, final exam, and design project report. Students will gain an understanding of modern parallel computer architectures.

Uploaded by

yeswanth1234
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 2

1

NJIT
ECE690: COMPUTER SYSTEMS ARCHITECTURE
COURSE SYLLABUS
Fall 2014

Instructor: Professor S.G. Ziavras, [email protected], (973) 596-3462.
Office hours: By appointment (Graduate Studies Office, Fenster 140).
Description: This course deals with the design and performance evaluation of advanced/high-
performance computer systems. The emphasis is on microprocessors, chip-multiprocessors and
memory hierarchy design. Historical information is presented as well along with data storage and
low-power dissipation schemes. Special attention is paid to pipelining, ILP (instruction-level
parallelism), DLP (data-level parallelism) and TLP (thread-level parallelism) using hardware)
and/or software techniques to yield high performance.
Prerequisites: Undergraduate degree in Computer Engineering, or ECE 684 or equivalent.
Textbook: Computer Architecture: A Quantitative Approach, by J ohn L. Hennessy and David A.
Patterson, Morgan Kaufmann, 5
th
edition, 2011, ISBN: 9780123838728.
Lecture notes and other references: http://web.njit.edu/~ziavras/classes.htm

Learning outcomes:
Understand the inner workings and performance capabilities of advanced microprocessors.
An ability to evaluate hardware accelerators targeting at applications with substantial data-level
parallelism (DLP).
Learn software-driven techniques to match application requirements to available pipelined
hardware in order to obtain high performance.
An ability to estimate the static and dynamic power dissipation of given hardware modules.
An ability to design microprocessor-based systems by accounting for performance and power
dissipation.
An ability to anticipate hardware performance improvements based on established rules from
past experiences with computer technology.
Improve report-writing skills when presenting results for computer design and evaluation.
Learn the differences among multiscalar, superpipelined, multithreaded, simultaneous
multithreaded, vector, and multicore processors.
Understand the forces behind the computer industrys shift to multicore processors.
Understand cache coherence issues.
An ability to design advanced memory hierarchies.
Understand the basic differences between shared-memory and message-passing interprocessor
connection networks.
An ability to select appropriate computer systems for given application domains.
Understand what hardware and software problems will require solutions for future generations
of multicore processors targeting at thread-level parallelism (TLP) and heterogeneous systems.


2

COURSE OUTLI NE
Weeks Chapter
1-3
Chapter 1: Fundamentals of Quantitative Design and Analysis
(Sections 1.1-1.9)
Appendix A: Instruction Set Principles (Sections A.1-A.10)
4-5

Appendix C: Pipelining: Basic and Intermediate Concepts
(Sections C.1-C.5)
6-7
Appendix B: Review of Memory Hierarchy (Sections B.1-B.3)
Chapter 2: Memory Hierarchy Design (Section 2.1)
8
MIDTERM EXAM
9-10
Chapter 3: Instruction-Level Parallelism and Its Exploitation
(Sections 3.1-3.12)
11-12

Chapter 5: Thread-Level Parallelism
(Sections 5.1-5.6)
13-14
Chapter 4: Data-Level Parallelism in Vector, SIMD, and GPU
Architectures (Sections 4.1-4.4)
Week 14: Project report due


Grading Policy:
Midterm exam: 35% (2 hours, open book, open notes)
Final exam 35% (2 hours, open book, open notes)
Project: 30% (report due in week 14)


The NJIT Honor Code will be upheld. Any violations will be brought to the immediate attention
of the Dean of Students. Please, access http://www.njit.edu/academics/pdf/academic-integrity-
code.pdf for NJ ITs academic integrity code.

You might also like