[PPT] - Intro duction 1 Objectives of this Lecture Course PowerPoint Presentation

SLIDE 1 Intro duction 1 Objectives

f

this Lecture

Course

Objectives

Course

Contents

Course

fo rmat, grading etc.

Why

distributed systems ?

Puzzles

c Vija y K. Ga rg Distributed Systems F all 94

SLIDE 2 Intro duction 2 Course Objective

Cover

fundamental concepts

Details
f

my resea rch interests

Ho

w to do resea rch in dist. systems ?

Mathematica

l to

ls
Mak

e friends

Instructo

r: Vija y Ga rg, ENS 527, 471-9424

email:

ga rg@ece.utexas.ed u

Oce

Hourse: TTh 3:30

5:00
Have

fun

solve

puzzles c Vija y K. Ga rg Distributed Systems F all 94

SLIDE 3 Intro duction 3 Course Contents

Global

Time

Global

State and Kno wledge

Algo

rithms: Mutual Exclusion, Causal Ordering

F

ault-tolerance

Distributed

Op erating Systems

Securit

y c Vija y K. Ga rg Distributed Systems F all 94

SLIDE 4 Intro duction 4 Course F

rmat
Grading
25

% Assignments

20

% Mid-T erm Exam

25

% T erm P ap er

30

% Final

Lectures
Questions
F

eedback c Vija y K. Ga rg Distributed Systems F all 94

SLIDE 5 Intro duction 5 Distributed Systems

Essential

cha racteristics

f

distributed systems

no

sha red clo ck

no

sha red memo ry 6 ?

I@

@R San F rancisco Austin New Y

rk

Comm uni cation Net w

rk

Time 12:01 Time 12:04 PE MEM MEM PE MEM PE Time 11:58 c Vija y K. Ga rg Distributed Systems F all 94

SLIDE 6 Intro duction 6 P a rallel Systems

Distributed

systems vs P a rallel systems

Physical

vs logical mo del P arallel System Distributed System Soft w are La y er Soft w are La y er e.g. D ASH e.g. PVM Distributed System P arallel System c Vija y K. Ga rg Distributed Systems F all 94

SLIDE 7 Intro duction 7 Advantages

f

Distributed Systems

Scalabilit

y

Sha

ring

f

resources

F

ault-tolerance

Ease

in p rogramming

Puzzle-lik

e qualit y c Vija y K. Ga rg Distributed Systems F all 94

SLIDE 8 Intro duction 8 Disadvantages

f

Distributed Systems ? ? ? Overhead Communication Synchronization Programming c Vija y K. Ga rg Distributed Systems F all 94

SLIDE 9 Intro duction 9 Existing Machines System F eatures In tel P aragon XP/S nCUBE/2 6480 P arsys Ltd. J Mac hine/MIT Sup erNo de1000 No de T yp e and 50 MHz i860 XP CISC 64-bit CPU m ultiple T-800 Message-Driv en Memory no des with 16-128 with FPU, 14 DMA T ransputers/no de Pro cessor Mb ytes/no de 1-64 Mb ytes/no de Net w

rk

and 2-D Mesh with SCSI 13-dimensional Recongurable 8

8
8

Mesh I/O HIPPI, VME, Ethernet Hyp ercub e

f

8192 in terconnect custom I/O no des OS 4.3 BSD V ertex/OS

r

UNIX IDRIS/OS w

rmhole

routing UNIX compatible Application Sparse Matrices Scien tic and Scien tic and Academic database academic P erformance 5-300 Gops p eak 27 Gop, 36 200 MIPS to 13 64-bit results Gb ytes/s I/O GIPS p eak Source: Adv anced Computer Arc h. b y Kai Hw ang and 1993 IEEE c Vija y K. Ga rg Distributed Systems F all 94

SLIDE 10 Intro duction 10 Time I disco v ered cure to AIDS I disco v ered cure to AIDS Earthling : Prof Ac heb e Martian : Prof T

cold

Time Barrier

clo

cks not synchronized

Ho

w to dene happ ed b efo re Time considered dangerous : replaced b y causalit y Reference: Lamp

rt

79 c Vija y K. Ga rg Distributed Systems F all 94

SLIDE 11 Intro duction 11 State

T

aking picture

f

sky

T

aking census in a country

Counting

numb er

f

tok ens in a distributed system

.

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

.

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

.

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

.

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

.

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

.

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

.

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . x Notion

f

physical state replaced b y consistent state Reference: Chandy and Lamp

rt

85 c Vija y K. Ga rg Distributed Systems F all 94

SLIDE 12 Intro duction 12 Secrets

Alice

calls up Bob.

Alice

and Bob do not sha re any p rivate info rmation.

Ho

w can Alice transmit a secret ?

Ho

w to sign y

ur

email messages ? Alice Eve Bob

@I

@R

?

Reference: Riv est, Shamir and Adleman 78 c Vija y K. Ga rg Distributed Systems F all 94

SLIDE 13 Intro duction 13 Kno wledge

F

ather: at least

ne
f

y

u

have mud

n

y

ur

fo rhead

He

rep eatedly asks the question: Do y

u

kno w if y

u

have mud

n

y

ur

fo rhead ?

What

happ ens ? F ather t u n children, k have mud Kno wing is dierent from kno wing what is kno wn Reference: Hailp ern and Moses 84 c Vija y K. Ga rg Distributed Systems F all 94

SLIDE 14 Intro duction 14 Ho w to check violation

f

Mutual Exclusion

A

company buys a single cop y

f

y

ur

p rogram.

can

run the p rogram

n

multiple machines

so

long as there is a single cop y

f

the p rogram at any time.

Ho

w will y

u

detect any violation

f

the agreement

allo

w ed access to a single computer at any time ? 6 ?

I@

@R v v San F rancisco Austin New Y

rk

Comm uni cation Net w

rk

Reference: Garg and W aldec k er 94 c Vija y K. Ga rg Distributed Systems F all 94

SLIDE 15 Intro duction 15 This course will not deal with..

Ha

rdw a re issues

Net

w

rking

issues

P

a rallel Algo rithms

Numerical

Metho ds c Vija y K. Ga rg Distributed Systems F all 94