Download CS116-Computer Architecture - Electrical and Computer Engineering

CENG 450
Computer Systems & Architecture
Lecture 1
Amirali Baniasadi
amirali@ece.uvic.ca
CENG 450: Computer Architecture
Instructor:
Amirali Baniasadi
EOW 441, Only by appt. Call or email with your schedule.
Email: amirali@ece.uvic.ca Office Tel: 721-8613
Web Page for this class will be at
http://www.ece.uvic.ca/~amirali/courses/ceng450.html
Text:
Computer Architecture A Quantitative Approach
Filth edition,
by Patterson and Hennessy
Lecture notes will be posted on the course web page in advance.
2
Course Structure
 Lectures:
 1 week on Overview and Introduction (Chap 1)
 2 weeks on ISA Design (Chap 2)
 6 weeks on Proc. Design (Chap 3 ,4)
 4 weeks on Memory and I/O (Chap 5)
 Reading assignments posted on the web for each week.
 NO Homework: Problems will be posted on the web site so you
can prepare for exams/quizzes.
 Quizzes: 4 in class quizzes. Dates will be announced in advance.
 Note that the above is approximate.
3
Course Philosophy
 Book to be used as supplement for lectures (If a topic is not covered in the
class, or a detail not presented in the class, that means I expect you to read
on your own to learn those details)
 One Project (25%)
 Four Quizzes (25%)- Will be announced in advance.
 Final Exam(50%)
 IMPORTANT NOTE: Must get passing grade in all components to
pass the course. Failing any of the three components will result in
failing the course.
4
Project
 Labs start at Week of Jan 23rd.
 Processor design.
5
Topics







Computer Architecture?
History
Technology
Moore’s law & Virtuous circle
Language evolution
Components of a computer
Instruction set architecture (ISA)
6
How many “computers” do you have?
 Three different computing markets:
1.Desktop Computing: low-end systems, high
performance workstations. Price $500 to $5000
2.Servers: web servers. Should be available and reliable.
Availability: be ready if components fail.
Scalability: ability to grow
3.Embedded computers: Hidden computers, ex. cell phones,
washing machine, palmtop, watch…
Minimize memory and power. Often not programmable.
7
What is “Computer Architecture”
Computer Architecture: Behind the doors!
Computer Architecture =
Instruction Set Architecture +
Machine Organization + Hardware
Instruction Set Architecture:
Visible to Compiler.
RISC vs. CISC.
Machine Organization: Importance of Von Newman design.
8
ISA
 1950s: Hardwired Control, easy to implement, limited resources
 1960s: Microprogramming, more flexibility.
 1970s: CISC:

Compilers in infancy so ISA designed for programmers.

Expensive & small memory: Highly encoded, Multiple size
instructions (e.g., x86 from 1-17 bytes), ISA approximates high
level languages,
 1980s: RISC:

Better compiler, cheaper memory, “elemental instructions”
 2000s: More resources, post-RISC?
CISC:”walk-across-the-room-without-stepping-on-the-dog”
RISC:”walk-walk-walk-step over dog-walk-walk”
9
History
1. “Big Iron” Computers:
Used vacuum tubes, electric relays and bulk magnetic storage
devices. No microprocessors. No memory.
Example: ENIAC (1945), IBM Mark 1 (1944)
10
History
Von Newmann:
Invented EDSAC (1949).
First Stored Program Computer.
Uses Memory.
Importance: We are still using the same basic design.
11
Computer Components
Output
Processor
Control
(CPU)
Memory
Input
Printer
Screen
Disk
...
keyboard
Mouse
Disk
...
12
Computer Components
 Datapath of a von Newman machine
OP1 + OP2
...
Op1
Op2
Bus
Op1
General-purpose
Registers
Op2
ALU i/p registers
ALU
OP1 + OP2
ALU o/p register
13
Computer Components
 Processor(CPU):
 Active part of the motherboard
 Performs calculations & activates devices
 Gets instruction & data from memory
 Components are connected via Buses
 Bus:
 Collection of parallel wires
 Transmits data, instructions, or control signals
 Motherboard
 Physical chips for I/O connections, memory, & CPU
14
Computer Components

CPU consists of
 Datapath (ALU+ Registers):
 Performs arithmetic & logical operations
 Control (CU):
 Controls the data path, memory, & I/O devices
 Sends signals that determine operations of datapath,
memory, input & output
15
Technology Change
 Technology changes rapidly
 HW
Vacuum tubes: Electron emitting devices
 Transistors: On-off switches controlled by electricity
Integrated Circuits( IC/ Chips): Combines thousands of transistors
Very Large-Scale Integration( VLSI): Combines millions of transistors
What next?
 SW
Machine language: Zeros and ones
Assembly language: Mnemonics
High-Level Languages: English-like
Artificial Intelligence languages: Functions & logic predicates
Object-Oriented Programming: Objects & operations on objects
16
Moore’s Prediction
17
Moore’s Law:
A new generation of memory chips is introduced every 3 years
Each new generation has 4 times as much memory as its predecessor
 Computer technology doubles every 1.5 years:
Example: DRAM capacity
100,000
64M
16M
K b it c a p a cit y


10,000
4M
1M
1000
256K
100
64K
16K
10
1976
1978
1980
1982
1984
1986
1988
Year o f introduction
1990
1992
1994
1996
18
Technology => dramatic change
Processor
logic capacity: about 30% per year
clock rate:
about 20% per year
Memory
DRAM capacity: about 60% per year (4x every 3 years)
Memory speed: about 10% per year
Cost per bit: improves about 25% per year
Disk
capacity: about 60% per year
Question: Does every thing look OK?
20
Software Evolution.




Machine language
Assembly language
High-level languages
Subroutine libraries
 There is a large gap between what is convenient for computers & what is
convenient for humans
 Translation/Interpretation is needed between both
21
Language Evolution
swap (int v[], int k)
{ int temp
temp = v[k];
v[k] = v[k+1];
v[k+1] = temp;
}
swap:
muli $2, $5, 4
add $2, $4, $2
lw
$15, 0($2)
lw
$18, 4($2)
sw
$18, 0($2)
sw
$15, 4($2)
jr
$31
0 00 0 0 0 0 0 1 0 1 0 0 0 0 1 0 0 0 00 00 00 00 1 1 0 0 0
0 00 0 0 0 0 0 1 0 0 0 1 1 1 0 0 0 0 1 1 0 0 0 0 0 1 00 0 0 1
1 00 0 1 1 0 0 0 1 1 0 0 0 1 0 0 0 0 00 0 0 0 0 00 00 0 0 0
1 00 0 1 1 0 0 1 1 1 1 0 0 1 0 0 0 0 00 0 0 0 0 00 00 1 0 0
1 0 1 0 1 1 0 0 1 1 1 1 0 0 1 0 0 0 0 00 0 0 0 0 00 00 0 0 0
1 0 1 0 1 1 0 0 0 1 1 0 0 0 1 0 0 0 0 00 0 0 0 0 00 00 1 0 0
0 00 0 0 0 1 1 1 1 1 0 0 0 0 0 0 0 0 00 0 0 0 0 0 0 0 1 0 0 0
High-level language
program (in C)
Assembly language
program (for MIPS)
Binary machine language
program (for MIPS)
22
HW - SW Components
 Hardware
 Memory components
Registers
Register file
memory
Disks
 Functional components
Adder, multiplier, dividers, . . .
Comparators
Control signals
 Software
 Data
Simple
• Characters
• Integers
• Floating-point
• Pointers
Structured
• Arrays
• Structures ( records)
 Instructions
Data transfer
Arithmetic
Shift
Control flow
Comparison
. . .
23
Things You Will Learn
 Assembly language introduction/Review
 How to analyze program performance
 How to design processor components
 How to enhance processors performance (caches, pipelines, parallel
processors, multiprocessors)
24
The Processor Chip
25
Processor Chip Major Blocks
• Example: Intel Pentium
• Area: 91 mm2
• ~ 3.3 million transistors ( 1 million for cache memory)
Control
Data
cache
Instruction
cache
Bus
Integer
datapath
Branch
Floatingpoint
datapath
26
Memory
 Categories
 Volatile memory
Loses information when power is switched-off
 Non-volatile memory
Keeps information when power is switched-off
 Types
 Cache:
Volatile
Fast but expensive
Smaller capacity
Placed closer to the processor
 Main memory
Volatile
Less expensive
 More capacity
 Secondary memory
Nonvolatile
Low cost
Very slow
Unlimited capacity
27
Input-Output (I/O)
 I/O devices have the hardest organization
Wide range of speeds
Graphics vs. keyboard
Wide range of requirements
Speed
Standard
Cost . . .
Least amount of research done in this area
28
Our Primary Focus
 The processor (datapath and control)
 Implemented using millions of transistors
 Impossible to understand by looking at each transistor
 We need abstraction
Hides lower-level details to offer simple model at higher level
Advantages
• Intensive & thorough research into the depths
• Reveals more information
• Omits unneeded details
• Helps us cope with complexity
Examples of abstraction:
• Language hierarchy
• Instruction set architecture (ISA)
29
Instruction Set Architecture (ISA)
 Instruction set:
 Complete set of instructions used by a machine
 ISA:
 Abstract interface between the HW and lowest-level SW. It encompasses
information needed to write machine-language programs including
Instructions
Memory size
Registers used
. . .
30
Instruction Set Architecture (ISA)
 ISA is considered part of the SW
 Several implementations for the same ISA can exist
 Modern ISA’s:
 80x86/Pentium/K6, PowerPC, DEC Alpha, MIPS, SPARC, HP
 We are going to study MIPS
 Advantages:
 Different implementations of the same architecture
 Easier to change than HW
 Standardizes instructions, machine language bit patterns, etc.
 Disadvantage:
 Sometimes prevents using new innovations
31
Instruction Set Architecture (ISA)
•Instruction Execution Cycle
Fetch Instruction From Memory
Decode Instruction determine its size & action
Fetch Operand data
Execute instruction & compute results or status
Store Result in memory
Determine Next Instruction’s address
32
What Should we Learn?
 A specific ISA (MIPS)
 Performance issues - vocabulary and motivation
 Instruction-Level Parallelism
 How to Use Pipelining to improve performance
 Exploiting Instruction-Level Parallelism w/ Software Approach
 Memory: caches and virtual memory
 I/O
33
What is Expected From You?
•
•
•
•
Read textbook & readings!
Be up-to-date!
Come back with your input & questions for discussion!
Appreciate and participate in teamwork!
34