Computer Systems Architecture A Networking Approach

Preface xiii
Preface to the first edition xv
Recommended lab sessions xxi
Part 1 Basic functions and facilities of a computer
Introduction: the hardware–software interface 3
1.1 Computer systems – the imprtance of networking 4
1.2 Hardware and software – mutual dependence 5
1.3 Programming your way into hardware – VHDL, a language for
electronic engineers 6
1.4 Systems administration – we all need to know 9
1.5 Voice, image and data – technological convergence 9
1.6 Windowing interfaces – WIMPs 11
1.7 The global Internet – connecting all the networks 13
1.8 Using the PC – a case study; more reasons to study CSA 16
The von Neumann Inheritance 23
2.1 Base 2 – the convenience of binary – 10110011100011110000 24
2.2 Stored program control – general-purpose machines 24
2.3 Instruction codes – machine action repertoire 26
2.4 Translation – compilers and assemblers 28
2.5 Linking – bringing it all together 28
2.6 Interpreters – executing high-level commands 30
2.7 Code sharing and reuse – let’s not write it all again! 31
2.8 Data codes – numeric and character 32
2.9 The operating system – Unix and Windows 36
2.10 Client–server computing – the way of the Net 40
2.11 Reconfigurable hardware – an alternative to fetch–execute 42
Functional units and the fetch–execute cycle 47
3.1 The naming of parts – CPU, memory, IO units 48
3.2 The CPU fetch–execute cycle – high-speed tedium 52
3.3 System bus – synchronous or asynchronous? 56
3.4 System clock – instruction cycle timing 59
3.5 Pre-fetching – early efforts to speed things up 61
3.6 Memory length – address width 63
3.7 Endian-ness – Microsoft vs. Unix, or Intel vs. Motorola? 65
3.8 Simple input–output – parallel ports 67
Building computers from logic: the control unit 73
4.1 Electronic Lego and logic – the advantage of modular units 74
4.2 Basic logic gates – truth tables for AND, OR, XOR and NOT 75
4.3 Truth tables and multiplexers – a simple but effective design tool 77
4.4 Programmable logic – reconfigurable logic chips 79
4.5 Traffic light controllers – impossible to avoid! 82
4.6 Circuit implementation from truth tables – some practical tips 83
4.7 Decoder logic – essential for control units and memories 85
4.8 CPU control unit – the ‘brain’ 87
4.9 Washing machine controllers – a simple CU 88
4.10 RISC vs. CISC decoding – in search of faster computers 91
Building computers from logic: the ALU 97
5.1 De Morgan’s equivalences – logical interchangeability 98
5.2 Binary addition – half adders, full adders, parallel adders 98
5.3 Binary subtraction – using two’s complement integer format 101
5.4 Binary shifting – barrel shifter 103
5.5 Integer multiplication – shifting and adding 105
5.6 Floating-point numbers – from very, very large to very, very small 108
Building computers from logic: the memory 117
6.1 Data storage – one bit at a time 118
6.2 Memory devices – memory modules for computers 120
6.3 Static memory – a lot of fast flip-flops 121
6.4 Dynamic memory – a touch of analogue amid the digital 122
6.5 DRAM refreshing – something else to do 124
6.6 Page access memories – EDO and SDRAM 124
6.7 Memory mapping – addressing and decoding 127
6.8 IO port mapping – integration vs. differentiation 131
The Intel Pentium CPU 137
7.1 The Pentium – a high-performance microprocessor 138
7.2 CPU registers – temporary store for data and address variables 143
7.3 Instruction set – introduction to the basic Pentium set 148
7.4 Structure of instructions – how the CU sees it 149
7.5 CPU status flags – very short-term memory 151
7.6 Addressing modes – building effective addresses 153
7.7 Execution pipelines – the RISC speedup technique 155
7.8 Pentium 4 – extensions 157
7.9 Microsoft Developer Studio – using the debugger 158
Subroutines 167
8.1 The purpose of subroutines – saving space and effort 168
8.2 Return address – introducing the stack 169
8.3 Using subroutines – HLL programming 170
8.4 The stack – essential to most operations 172
8.5 Passing parameters – localizing a subroutine 173
8.6 Stack frame – all the local variables 176
8.7 Supporting HLLs – special CPU facilities for dealing with subroutines 179
8.8 Interrupt service routines – hardware-invoked subroutines 179
8.9 Accessing operating system routines – late binding 180
Simple input and output 185
9.1 Basic IO methods – polling, interrupt and DMA 186
9.2 Peripheral interface registers – the programmer’s viewpoint 187
9.3 Polling – single-character IO 191
9.4 Interrupt processing – service on demand 197
9.5 Critical data protection – how to communicate with interrupts 205
9.6 Buffered IO – interrupt device drivers 209
9.7 Direct memory access (DMA) – autonomous hardware 210
9.8 Single-character IO – screen and keyboard routines 212
Serial Connections 219
10.1 Serial transmission – data, signals and timing 220
10.2 Data format – encoding techniques 221
10.3 Timing synchronization – frequency and phase 224
10.4 Data codes and error control – parity, checksums, Hamming codes
and CRCs 227
10.5 Flow control – hardware and software methods 235
10.6 The 16550 UART – RS232 237
10.7 Serial mice – mechanical or optical 244
10.8 Serial ports – practical tips, avoiding the frustration 246
10.9 USB – Universal Serial Bus 246
10.10 Modems – modulating carrier waves 252
Parallel connections 263
11.1 Parallel interfaces – better performance 264
11.2 Centronics – more than a printer port but less than a bus 264
11.3 SCSI – the Small Computer Systems Interface 267
11.4 IDE – Intelligent Drive Electronics 271
11.5 AT/ISA – a computer standards success story 272
11.6 PCI – Peripheral Component Interconnection 275
11.7 Plug-and-Play – automatic configuration 278
11.8 PCMCIA – Personal Computer Memory Card
International Association 280
The memory hierarchy 285
12.1 Levels of performance – you get what you pay for 286
12.2 Localization of access – exploiting repetition 288
12.3 Instruction and data caches – matching memory to CPU speed 293
12.4 Cache mapping – direct or associative 295
12.5 Virtual memory – segmentation and demand paging 299
12.6 Address formulation – when, where and how much 304
12.7 Hard disk usage – parameters, access scheduling and
data arrangement 306
12.8 Performance improvement – blocking, caching, defragmentation,
scheduling, RAM disk 310
12.9 Optical discs – CD-DA, CD-ROM, CD-RW and DVDs 312
12.10 DVD – Digital Versatile Disc 316
12.11 MPEG – video and audio compression 316
12.12 Flash sticks – the new floppy disk 323
Part 2 Networking and increased complexity
The programmer’s viewpoint 329
13.1 Different viewpoints – different needs 330
13.2 Application user – office packages 331
13.3 Systems administration – software installation and maintenance 333
13.4 HLL programmer – working with Java, C++, BASIC or C# 337
13.5 Systems programming – assembler and C 340
13.6 Hardware engineer – design and hardware maintenance 344
13.7 Layered virtual machines – hierarchical description 345
13.8 Assemblers – simple translators 346
13.9 Compilers – translation and more 347
Local area networks 353
14.1 Reconnecting the users – email, printers and database 354
14.2 PC network interface – cabling and interface card 359
14.3 Ethernet – Carrier Sense, Multiple Access/Collision Detect 363
14.4 LAN addressing – logical and physical schemes 367
14.5 Host names – another layer of translation 370
14.6 Layering and encapsulation – TCP/IP software stack 371
14.7 Networked file systems – sharing files across a network 372
14.8 Interconnecting networks – gateways 374
14.9 Socket programming – an introduction to WinSock 374
Wide area networks 383
15.1 The Internet – origins 384
15.2 TCP/IP – the essential protocols 386
15.3 TCP – handling errors and flow control 390
15.4 IP routing – how packets find their way 392
15.5 DNS – Distributed Name Database 398
15.6 World Wide Web – the start 401
15.7 Browsing the Web – Netscape Navigator 403
15.8 HTTP – another protocol 407
15.9 Search engines – Google 409
15.10 Open Systems Interconnect – an idealized scheme 412
Other networks 419
16.1 The PSTN – telephones 420
16.2 Cellnets – providers of mobile communications 426
16.3 ATM – Asynchronous Transfer Mode 435
16.4 Messaging – radio paging and packet radio networks 440
16.5 ISDN – totally digital 442
16.6 DSL – Digital Subscriber Line 446
16.7 Cable television – facilities for data transmission 447
Introduction to operating systems 455
17.1 Historic origins – development of basic functions 456
17.2 Unix – a landmark operating system 459
17.3 Outline structure – modularization 462
17.4 Process management – initialization and dispatching 463
17.5 Scheduling decisions – time-slicing, demand preemption
or cooperative 469
17.6 Task communication – pipes and redirection 471
17.7 Exclusion and synchronization – semaphores and signals 473
17.8 Memory allocation – malloc( ) and free( ) 479
17.9 User interface – GUIs and shells 481
17.10 Input–output management – device handlers 482
Windows XP 491
18.1 Windows GUIs – responding to a need 492
18.2 Win32 – the preferred user API 494
18.3 Processes and threads – multitasking 495
18.4 Memory management – virtual memory implementation 496
18.5 Windows Registry – centralized administrative database 496
18.6 NTFS – Windows NT File System 498
18.7 File access – ACLs, permissions and security 499
18.8 Sharing software components – OLE, DDE and COM 502
18.9 Windows NT as a mainframe – Winframe terminal server 502
Filing systems 507
19.1 Data storage – file systems and databases 508
19.2 The PC file allocation table – FAT 515
19.3 Unix inodes – they do it differently 518
19.4 Microsoft NTFS – complexity and security 523
19.5 RAID configuration – more security for the disk subsystem 525
19.6 File security – access controls 526
19.7 CD portable file system – multi-session contents lists 528
Visual output 533
20.1 Computers and graphics – capture, storage, processing
and redisplay 534
20.2 PC graphics adapter cards – graphics coprocessors 541
20.3 Laser printers – this is mechatronics! 547
20.4 Adobe PostScript – a page description language 549
20.5 WIMPs – remodelling the computer 554
20.6 Win32 – graphical API and more 555
20.7 The X Window system – enabling distributed processing 557
20.8 MMX technology – assisting graphical calculations 558
RISC processors: ARM and SPARC 563
21.1 Justifying RISC – increased instruction throughput 564
21.2 Pipeline techniques – more parallel operations 569
21.3 Superscalar methods – parallel parallelism 571
21.4 Register files – many more CPU registers 572
21.5 Branch prediction methods – maintaining the pipelines 574
21.6 Compiler support – an essential part of RISC 576
21.7 The ARM 32 bit CPU – origins 576
21.8 StrongARM processor – a 32 bit microcontroller 585
21.9 The HP iPAQ – a StrongARM PDA 588
21.10 Puppeteer – a StrongARM SBC 590
21.11 Sun SPARC – scalar processor architecture as RISC 592
21.12 Embedded systems – cross-development techniques 594
VLIW processors: the EPIC Itanium 601
22.1 Itanium 64 bit processor – introduction 602
22.2 Itanium assembler – increasing the control of the CPU 609
22.3 Run-time debugging – gvd/gdb 613
22.4 Future processor design – debate 615
Parallel processing 619
23.1 Parallel processing – the basis 620
23.2 Instruction-level parallelism (ILP) – pipelining 623
23.3 Superscalar – multiple execution units 623
23.4 Symmetric, shared memory multiprocessing (SMP) – the future? 623
23.5 Single-chip multiprocessors – the IBM Cell 626
23.6 Clusters and grids – application-level parallelism 629
Appendix: MS Visual Studio 8, Express Edition 635
Glossary 647
Answers to end-of-chapter questions 661
References 713
Index 717

Computer architecture - Wikipedia, the free encyclopedia, EE282 - Computer Systems Architecture - Stanford University

A Networking Approach to Grid Computing
Data Warehousing and Data Mining for Telecommunications
Other Core of CS Books
Other Network Books