ECE 56500 - Computer Architecture

Lecture Hours: 3 Credits: 3

Counts as:
CMPE Special Content Elective
EE Elective

Normally Offered: Each Fall

Requisites:
ECE 43700

Catalog Description:
Computer architecture is the science and art of selecting and interconnecting hardware components to create a computer that meets functional, performance and cost goals. This course qualitatively and quantitatively examines uniprocessor computer design trade-offs. We will learn, for example, how uniprocessors execute many instructions concurrently and why state-of-the-art memory systems are nearly as complex as processors. Examining tradeoffs requires that you already know how to correctly design a computer, as is taught in the important prerequisite EE 43700.

Required Text(s):
  1. Computer Architecture - A Quantitative Approach, 5th Edition, D. Patterson and J. Hennessy, Elsevier Science, 2011, ISBN No. 978-0123838728.
Recommended Text(s):
  1. Readings in Computer Architecture, Mark Hill, Norman Jouppi and Gurindar Sohi, Morgan Kaufmann Publishers.

Learning Outcomes:

A student who successfully fulfills the course requirements will have demonstrated:
  1. an understanding of advanced processor pipelines including mechanisms for hardware or compiler instruction scheduling. renaming, branch prediction, and precise interrupts. [1,2,6]
  2. an understanding of advanced cache techniques including hardware and software schemes for miss ratio reduction and miss penalty reduction, and high-bandwidth design for cache and main memory. [1,2,6]
  3. an understanding of virtual memory including page table design, TLBs, and interaction between caches and address translation. [1,2,6]
  4. an understanding of basic snoopy coherence and the notion of memory consistency models. [1,2,6]

Lecture Outline:

Weeks Topic
0.5 Introduction
1.0 Performance and Cost
1 Pipelining - implementation and evaluation
4 Advanced pipelining - hardware and compiler techniques
1 Memory Hierarchy - cache design and evaluation
4 Advanced cache and compiler techniques, and virtual memory
1 Input/Output - devices and performance
1.5 Vector processing and multiprocessors
1 Exams