What this book covers
The information in this book is presented in the following sequence:
Chapter 1, Introducing Computer Architecture, begins with a brief history of automated computing devices and describes the significant technological advances that drove leaps in capability. This is followed by a discussion of Moore's law, with an assessment of its applicability over previous decades and the implications for the future. The basic concepts of computer architecture are introduced in the context of the 6502 microprocessor.
Chapter 2, Digital Logic, introduces transistors as switching elements and explains their use in constructing logic gates. We will then see how flip-flops and registers are developed by combining simple gates. The concept of sequential logic, meaning logic that contains state information, is introduced, and the chapter ends with a discussion of clocked digital circuits.
Chapter 3, Processor Elements, begins with a conceptual description of a generic processor. We will examine the concepts of the instruction set, register set, and instruction loading, decoding, execution, and sequencing. Memory load and store operations are also discussed. The chapter includes a description of branching instructions and their use in looping and conditional processing. Some practical considerations are introduced that lead to the necessity for interrupt processing and I/O operations.
Chapter 4, Computer System Components, discusses computer memory and its interface to the processor, including multilevel caching. I/O requirements including interrupt handling, buffering, and dedicated I/O processors are described. We will discuss some specific requirements for I/O devices including the keyboard and mouse, the video display, and the network interface. The chapter ends with descriptive examples of these components in modern computer applications, including smart mobile devices, personal computers, gaming systems, cloud servers, and dedicated machine learning systems.
Chapter 5, Hardware-Software Interface, discusses the implementation of the high-level services a computer operating system must provide, including disk I/O, network communications, and interactions with users. This chapter describes the software layers that implement these features starting at the level of the processor instruction set and registers. Operating system functions, including booting, multiprocessing, and multithreading, are also described.
Chapter 6, Specialized Computing Domains, explores domains of computing that tend to be less directly visible to most users, including real-time systems, digital signal processing, and GPU processing. We will discuss the unique requirements associated with each of these domains and look at examples of modern devices implementing these features.
Chapter 7, Processor and Memory Architectures, takes an in-depth look at modern processor architectures, including the von Neumann, Harvard, and modified Harvard variants. The chapter discusses the implementation of paged virtual memory. The practical implementation of memory management functionality within the computer architecture is introduced and the functions of the memory management unit are described.
Chapter 8, Performance-Enhancing Techniques, discusses a number of performance-enhancing techniques used routinely to reach peak execution speed in real-world computer systems. The most important techniques for improving system performance, including the use of cache memory, instruction pipelining, instruction parallelism, and SIMD processing, are the subjects of this chapter.
Chapter 9, Specialized Processor Extensions, focuses on extensions commonly implemented at the processor instruction set level to provide additional system capabilities beyond generic data processing requirements. The extensions presented include privileged processor modes, floating-point mathematics, power management, and system security management.
Chapter 10, Modern Processor Architectures and Instruction Sets, examines the architectures and instruction set features of modern processor designs including the x86, x64, and ARM processors. One challenge that arises when producing a family of processors over several decades is the need to maintain backward compatibility with code written for earlier-generation processors. The need for legacy support tends to increase the complexity of the later-generation processors. This chapter will examine some of the attributes of these processor architectures that result from supporting legacy requirements.
Chapter 11, The RISC-V Architecture and Instruction Set, introduces the exciting new RISC-V (pronounced risk five) processor architecture and its instruction set. RISC-V is a completely open source, free-to-use specification for a reduced instruction set computer architecture. A complete user-mode (non-privileged) instruction set specification has been released and a number of hardware implementations of this architecture are currently available. Work is ongoing to develop specifications for a number of instruction set extensions. This chapter covers the features and variants available in the RISC-V architecture and introduces the RISC-V instruction set. We will also discuss the applications of the RISC-V architecture in mobile devices, personal computers, and servers.
Chapter 12, Processor Virtualization, introduces the concepts involved in processor virtualization and explains the many benefits resulting from the use of virtualization. The chapter includes examples of virtualization based on open source tools and operating systems. These tools enable the execution of instruction-set-accurate representations of various computer architectures and operating systems on a general-purpose computer. We will also discuss the benefits of virtualization in the development and deployment of real-world software applications.
Chapter 13, Domain-Specific Computer Architectures, brings together the topics discussed in previous chapters to develop an approach for architecting a computer system design to meet unique user requirements. We will discuss some specific application categories, including mobile devices, personal computers, gaming systems, Internet search engines, and neural networks.
Chapter 14, Future Directions in Computer Architectures, looks at the road ahead for computer architectures. This chapter reviews the significant advances and ongoing trends that have resulted in the current state of computer architectures and extrapolates these trends in possible future directions. Potentially disruptive technologies are discussed that could alter the path of future computer architectures. In closing, I will propose some approaches for professional development for the computer architect that should result in a future-tolerant skill set.