Transcript Introduction

CMSC 611
Advanced Computer Arch.
Getting Started
Some material adapted from Mohamed Younis, UMBC CMSC 611 Spr 2003 course slides
Some material adapted from David Culler, UC Berkeley CS252, Spr 2002 course slides, © 2002 UC Berkeley
Some material adapted from Hennessy & Patterson / © 2003 Elsevier Science
2
Overview
•
•
•
•
•
•
•
Resources, syllabus, work load
Grade structure and policy
Expected background
An introduction to computer architecture
Why study computer architecture?
Organization and anatomy of computers
Impact of microelectronics technology on
computers
• The evolution of the computer industry
and generations
3
Course Resources
• Instructor: Marc Olano / ITE 354
– Office Hours: TuTh 4:00 – 5:00
• TA: Tiantian Xie / ITE 344
– Office Hours: W 4:00-6:00
• Web Page:
– www.umbc.edu/~olano/611
• Book
– Hennessy and Patterson, Computer
Architecture: A Quantitative Approach,
3rd, 4th, or 5th Edition
4
Syllabus
•
•
•
•
•
•
•
Quantitative Design Principles
Instruction Set Principles
Pipelining and Instruction Parallelism
Memory Hierarchy Design
Storage and I/O
Multiprocessor Systems
Interconnection Networks
5
Workload
• Assignments
– Approximately 2 hours, every other week
• Exams
– Midterm in class, Thursday, March 24th
– Final Thursday, May 12th, 10:30 – 12:30
• Project
6
Project
• Teams of 3-4
• You choose application area
– Best to choose your own research area
• Design architecture for your application
• Final written report / architecture manual
7
Grades
• Breakdown
– 30% Homework (5% each)
– 30% Project
– 20% Midterm
– 20% Final
• Homework policy
– Due in class, ZERO if late
– Exceptions allowed for documented work
or conference travel, approved in advance
8
Grades
•
•
•
•
•
•
•
•
•
•
A
AB+
B
BC+
C
CD
F
≥ 93
≥ 90
≥ 87
≥ 83
≥ 80
≥ 77
≥ 73
≥ 70
≥ 60
< 60
• No curving
– Not even half a point
– Expect some extra
credit assignments
• At least several %
– Get more practice to
get a better grade
9
Expected Background
• CMSC 411: Computer Architecture
– Design of computer systems
•
•
•
•
•
•
•
•
Information representation
Floating point arithmetic
Hardwired & micro programmed control
Pipelining
Cache
Bus control & timing
I/O mechanisms
Parallel processing
• 411 focus on design and implementation (how)
• We focus on design decisions (why)
10
Introduction & Motivation
• Even if you don’t want to do computer
architecture, this class will
– Help you understand the limits & capabilities CPUs
– Help you understand why
– Help you understand how to write better code
•
•
•
•
Extreme importance of cache coherence
Importance of branch predictability, or avoiding branching
Folly of micro-optimization that the CPU will undo
How to prioritize optimization work in hardware or software
– Tools of computer architecture apply everywhere!