A computer system consists of hardware, system programs, and application programs figs 8

PRINCIPLES OF I/O HARDWARE 5.2 PRINCIPLES OF I/O SOFTWARE 5.3 I/O SOFTWARE LAYERS 5.4 DISKS 5.5 CLOCKS 5.6 CHARACTER-ORIENTED TERMINALS 5.7 GRAPHICAL USER INTERFACES 5.8 NETWORK TERMINALS 5.9 POWER MANAGEMENT 5.10 RESEARCH ON INPUT/OUTPUT 5.11 SUMMARY

C C C C

M C C

C Shared

memory

connect

Inter-CPU

Local memory

M C C M

message-CPU CPU M

Shared memory

Shared memory

Fig 8-2 Three bus-based multiprocessors (a) Without caching.(b) With caching (c) With caching and private memories

Closed crosspoint switch

Open crosspoint switch (a)

(b)

(c)

Crosspoint switch is closed

Crosspoint switch is open 000

B

X Y

Module Address Opcode Value

Fig 8-4 (a) A 2 × 2 switch (b) A message format

0 0 1 0 0

CPU 2

Has private OS

CPU 3

Has private OS

1 2 Data Data

3 4 Data Data

OS code

Bus

Fig 8-7 Partitioning multiprocessor memory among four CPUs,but sharing a single copy of the operating system code The boxesmarked Data are the operating system’s private data for each CPU

CPU 2

Slave runs user processes

CPU 3

User processes OS

Bus

Slave runs user processes

Fig 8-8 A master-slave multiprocessor model

CPU 2

Runs users and shared OS

CPU 3

Runs users and

Locks Bus

Fig 8-9 The SMP multiprocessor model

CPU 1 Memory CPU 2

Bus

Word

1000 is initially 0

CPU 3

CPU 3 spins on this (private) lock

CPU 4 spins on this (private) lock CPU 2 spins on this (private) lock

When CPU 1 is finished with the real lock, it releases it and also releases the private lock CPU 2

is spinning on

CPU 1 holds the real lock Shared memory

4 2

3

1

Fig 8-11 Use of multiple locks to avoid cache thrashing

1 5 9 13

2 6 10 14

3 7 11 15

B C

D E F

G H I

J K

L M N

7 5 4 2 1 0

Priority

CPU 4 goes idle

CPU 12 goes idle

0 A 8 B

1 5 9 13

2 6 10 14

3 7 11 15

C

D E F

G H I

J K

L M N

7 5 4 2

1 0 Priority

Fig 8-12 Using a single data structure for scheduling a cessor

6-CPU partition

8-CPU partition

Fig 8-13 A set of 32 CPUs split into four partitions, with twoCPUs available

Fig 8-14 Communication between two threads belonging to

process A that are running out of phase.

CPU 1 Input port

(a)

Output port

Entire packet

Entire

packet

Four-port switch

C

A

CPU 2

Entire packet D

(c)

C

A

D B

Fig 8-17 Store-and-forward packet switching

CPU

CPU

CPU Switch

Node 2 Main RAM

Main RAM

Node 4

Interface board

Optional on- board CPU Interface

board RAM Node 3

Main RAM

Main RAM Node 1

3 2

multicom-Node 1 Node 2

CPU

7 2 3 4 6

Receive ring CPUSend ring

7 0 1 2 4

OS OS

Fig 8-19 Use of send and receive rings to coordinate the mainCPU with the on-board CPU

Sender blocked

Sender blocked

Trap to kernel, sender blocked

Message being sent

Message being sent Sender running

Sender running

Return Sender running

Sender running

Trap

Message copied to a kernel buffer

Return from kernel, sender released

(a)

(b)

Fig 8-20 (a) A blocking send call (b) A nonblocking send call

Client CPU

Client stub Client

2 1

Operating system

Server CPU Server

Network

Fig 8-21 Steps in making a remote procedure call The stubs areshaded gray

Operating system Application

Hardware

(b)

Run-time system

Operating system

Shared memory

Application

Hardware

Run-time system

Operating system Application

Hardware

(c)

Run-time system

Operating system

Shared memory

Application

Hardware

Run-time system

Operating system Application

Hardware

Fig 8-22 Various layers where shared memory can be mented (a) The hardware (b) The operating system (c) User-level software

imple-Globally shared virtual memory consisting of 16 pages

Memory

Network (a)

CPU 1

Code using

variable A

A B

Shared

page

CPU 2

Code using variable B

A B

Network

A and B are unrelated shared variables that just happen to be on the same page

Fig 8-24 False sharing of a page containing two unrelated ables

8 1 2

8 1 2

4

3 6 2

1

4

Node 3 Traffic

I’m full Here, have a process

Take some work

SPARC Solaris

Application

Mac OS Application

Macintosh Common base for applications

Network

Fig 8-28 Positioning of middleware in a distributed system

Ethernet

Switch Computer

Ethernet

(b) (a)

Fig 8-29 (a) Classic Ethernet (b) Switched Ethernet

Backbone High-bandwidth fiber

Reliable message stream Reliable byte stream Unreliable connection Unreliable datagram Acknowledged datagram Request-reply

Example

Sequence of pages of a book Remote login

Digitized voice Network test packets Registered mail Database query

Connection-oriented

Connectionless

Fig 8-31 Six different types of network service

Ethernet 1 header

School of Humanities School of Sciences School of Social Sciences

Science Astronomy Biology Chemistry Physics Main page

Social sciences Anthropology Psychology Sociology Main page

Astronomy Dept.

Galaxies Nebulas Planets Quasars Stars Sciences

Biology Dept.

Arachnids Protozoa Worms Sciences

Chemistry Dept.

Acids Bases Proteins Sciences

Physics Dept.

Electrons Mesons Neutrons Neutrinos Protons Sciences

Anthropology Dept.

African tribes Australian tribes New Guinean tribes Social sciences

Psychology Dept.

Freud Rats Social sciences

Sociology Dept Class struggle Gender struggle Generic struggle Social sciences

Fig 8-33 The Web is a big directed graph of documents

2 Accesses are

done on the

client

3 When client is done, file is returned to server

Fig 8-34 (a) The upload/download model (b) The remote accessmodel

Fig 8-35 (a) Two file servers The squares are directories and thecircles are files (b) A system in which all clients have the sameview of the file system (c) A system in which different clientsmay have different views of the file system.

Fig 8-36 (a) Sequential consistency (b) In a distributed systemwith caching, reading a file may return an obsolete value.

User process

Vice server

Operating system Operating system

Venus

Network

Symbolic link

Root directory bin

cmu cache etc lib tmp

cp ls sh

file 1 file 2 file 3

cell1 cell2 cell3 cell4

motd passwd

sh

Fig 8-37 (a) The position of venus and vice in AFS (b) A

client’s view of the file system

Client Client stub

Operating system Client ORB

Client

code

Object adapter

Operating system

Server code

Server Skeleton

IIOP protocol

Network Server ORB

Fig 8-38 The main elements of a distributed system based onCORBA The CORBA parts are shown in gray

List messages Read message Append message Delete message State of

Class object contains the method

Fig 8-39 The structure of a Globe object

R W

R W

R W

R W

Computer Object

Semantics subobject

Communication subobject

Interface Control

subobject

Replication

subobject

Security subobject

Operating system

Messages in and out go through the communication subobject Network

Fig 8-41 Structure of a Globe object

("abc", 2, 5)

("matrix-1", 1, 6, 3.14)

("family", "is-sister", "Stephany", "Roberta")

Fig 8-42 Three Linda tuples

Information router

Daemon Consumer