1 International Technology University CEN 951 Computer Architecture Lecture 1 - Introduction

2 Introduction Rapidly changing field: vacuum tube -> transistor -> IC -> VLSI Doubling every 1.5 years memory capacity processor speed (Due to advances in technology and organization)

3 Intel Note: 16 wires bonded lines in parallel are buses function modules are rectangular

4 Pentium IIIb 1997 Bond pads on border buses are parallel lines function blocks rectangular

5 Processor speed in MIPS Processor speed increases by factor 10 each 5 years

6 Growth of DRAM chip capacity DRAM capacity grows by factor 10 each 5 years

7 Things you’ll be learning: How computers work, a basic foundation How to analyze their performance (or how not to!) Issues affecting modern processors (caches, pipelines) Why learn this stuff? You want to call yourself a “computer scientist” You want to build software people use (need performance) You need to make a purchasing decision or offer “expert” advice You want to understand specs of current Pentium

8 What is a computer? Components: –input (mouse, keyboard) –output (display, printer) –memory (disk drives, DRAM, SRAM, CD) –network Our primary focus: –the processor (data- path and control) implemented using millions of transistors Impossible to understand by looking at each transistor We need…abstraction

9 PC motherboard in case with cooling for CPU, daughter boards for peripherals

10

11 7 Layer model for architecture High Level Low Level User Level: Application Programs High Level Languages Assembly Language / Machine Code Micro-programmed / Hardwired Control Functional Units (Memory, ALU, etc.) Logic Gates Transistors and Wires

12

13 Abstraction Delving into the depths reveals more information An abstraction –omits unneeded detail –helps us cope with complexity What are some of the details that appear in these familiar abstractions?

14 High- level language program (in C) and equivalent assembler code 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 $16, 4($2) sw $16, 0($2) sw $15, 4($2) jr $31 v[k] v[k+1] $5=k, $4=addr of v[0] return from routine

15 Assembly language program (for MIPS), and binary machine language program (for MIPS) swap: muli $2, $5, 4 add $2, $4,$ 2 lw $15, 0($ 2) lw $16, 4($ 2) sw $16, 0($ 2) sw $15, 4($ 2) jr $ These 32 bit codes are not exactly the MIPS codes!!

16

17 Instruction Set Architecture A very important abstraction –interface between hardware and low- level software –standardizes instructions, machine language bit patterns, etc. advantage: –different implementations of the same architecture disadvantage: –sometimes prevents using new innovations

18 Modern instruction set architectures 80x86/ Pentium/ K6, PowerPC, DEC Alpha, MIPS, SPARC, HP

19 Where we are headed Performance issues, vocabulary and motivation A specific instruction set architecture Arithmetic and how to build an ALU Constructing a processor to execute our instructions Pipelining to improve performance Memory: caches and virtual memory I/ O Assembler programming

20 Desirable features of systems coherent collaborative connected extensible flexible hidden instrumented maintainable open reusable robust scalable secure transportable

21 coherent made of individual components with defined interfaces, unaffected by changes in implementation of other components

22 collaborative data and transactions work across environments from different vendors and implementations

23 connected information and functionality are available by defined interfaces, paths and connections from elsewhere in the system

24 extensible the system components can be extended to new unanticipated contexts and situations

25 flexible performance degrades gradually allowing time for corrective action with evolving business context and environment

26 hidden data interfaces are well defined, complexity of implementation is hidden, simplicity of design is presented

27 instrumented the system is provided with built in sensors or data gathering devices for system diagnostics

28 maintainable useful life of the system can be prolonged by routine attention to features that change or wear out

29 open implemented in publicly available standards

30 reusable components are well defined, subjected to configuration management, and documented so that they can be used again in different contexts

31 robust able to adapt to changing requirements and contexts, allowing for effective modification, administration and management; gradual degredation under stress

32 scalable performance improves linearly or nearly linearly as system components are added to handle additional data volume, users and processing requirements

33 secure defensible against intentional or accident attack or damage

34 transportable components of the system can be installed in a variety of implementations Also consider: –Security –Availability –Performance