Book Image

Modern Computer Architecture and Organization

By : Jim Ledin
Book Image

Modern Computer Architecture and Organization

By: Jim Ledin

Overview of this book

Are you a software developer, systems designer, or computer architecture student looking for a methodical introduction to digital device architectures but overwhelmed by their complexity? This book will help you to learn how modern computer systems work, from the lowest level of transistor switching to the macro view of collaborating multiprocessor servers. You'll gain unique insights into the internal behavior of processors that execute the code developed in high-level languages and enable you to design more efficient and scalable software systems. The book will teach you the fundamentals of computer systems including transistors, logic gates, sequential logic, and instruction operations. You will learn details of modern processor architectures and instruction sets including x86, x64, ARM, and RISC-V. You will see how to implement a RISC-V processor in a low-cost FPGA board and how to write a quantum computing program and run it on an actual quantum computer. By the end of this book, you will have a thorough understanding of modern processor and computer architectures and the future directions these architectures are likely to take.
Table of Contents (20 chapters)
1
Section 1: Fundamentals of Computer Architecture
8
Section 2: Processor Architectures and Instruction Sets
14
Section 3: Applications of Computer Architecture

Neural networks and machine learning architectures

We briefly reviewed the architecture of neural networks in Chapter 6, Specialized Computing Domains. This section examines the inner workings of a high-performance, dedicated neural net processor.

Intel Nervana neural network processor

In 2019, Intel announced the release of a pair of new processors, one optimized for the task of training sophisticated neural networks and the other for using trained networks to conduct inference, which is the process of generating neural network outputs given a set of input data.

The Nervana neural network processor for training (NNP-T) is essentially a miniature supercomputer tailored to the computational tasks required in the neural network training process. The NNP-T1000 is available in the following two configurations:

  • The NNP-T1300 is a dual-slot PCIe card suitable for installation in a standard PC. It communicates with the host via PCIe 3.0 or 4.0 x16. It is possible to connect...