Modern Processor Design

Modern Processor Design
Author: John Paul Shen
Publisher: Waveland Press
Total Pages: 657
Release: 2013-07-30
Genre: Computers
ISBN: 147861076X

Download Modern Processor Design Book in PDF, Epub and Kindle

Conceptual and precise, Modern Processor Design brings together numerous microarchitectural techniques in a clear, understandable framework that is easily accessible to both graduate and undergraduate students. Complex practices are distilled into foundational principles to reveal the authors insights and hands-on experience in the effective design of contemporary high-performance micro-processors for mobile, desktop, and server markets. Key theoretical and foundational principles are presented in a systematic way to ensure comprehension of important implementation issues. The text presents fundamental concepts and foundational techniques such as processor design, pipelined processors, memory and I/O systems, and especially superscalar organization and implementations. Two case studies and an extensive survey of actual commercial superscalar processors reveal real-world developments in processor design and performance. A thorough overview of advanced instruction flow techniques, including developments in advanced branch predictors, is incorporated. Each chapter concludes with homework problems that will institute the groundwork for emerging techniques in the field and an introduction to multiprocessor systems.

Instruction Level Parallel Processors -- a New Architectural Model for Simulation and Analysis

Instruction Level Parallel Processors -- a New Architectural Model for Simulation and Analysis
Author: Stanford University. Computer Systems Laboratory
Publisher:
Total Pages: 31
Release: 1994
Genre: Computer architecture
ISBN:

Download Instruction Level Parallel Processors -- a New Architectural Model for Simulation and Analysis Book in PDF, Epub and Kindle

Trends in high-performance computer architecture have led to the development of increased clock-rate and dynamic multiple-instruction issue processor designs. There have been problems combining both these techniques due to the pressure that the complex scheduling and issue logic puts on the cycle time. This problem has limited the performance of multiple-instruction issue architectures. The alternative approach of static multiple-operation issue avoids the clock-rate problem by allowing the hardware to concurrently issue only those operations that the compiler scheduled to be issued concurrently. Since there is no hardware support required to achieve multiple-operation issue (there are multiple operations in a single instruction and the hardware issues a single instruction at a time), these designs can be effectively scaled to high clock rates. However, these designs have the problem that the scheduling of operations into instructions is rigid and to increase the performance of the system the entire system must be scaled uniformly so that the static schedule is not compromised. This report describes an architectural model that allows a range of hybrid architectures to be studied.

Modern Processor Design: Fundamentals of Superscalar Processors

Modern Processor Design: Fundamentals of Superscalar Processors
Author: John Shen
Publisher: McGraw-Hill Science/Engineering/Math
Total Pages: 656
Release: 2004-07-07
Genre: Technology
ISBN: 9780070570641

Download Modern Processor Design: Fundamentals of Superscalar Processors Book in PDF, Epub and Kindle

Modern Processor Design: Fundamentals of Superscalar Processors is an exciting new first edition from John Shen of Carnegie Mellon University & Intel and Mikko Lipasti of the University of Wisconsin--Madison. This book brings together the numerous microarchitectural techniques for harvesting more instruction-level parallelism (ILP) to achieve better processor performance that have been proposed and implemented in real machines. Other advanced techniques from recent research efforts that extend beyond ILP to exploit thread-level parallelism (TLP) are also compiled in this book. All of these techniques, as well as the foundational principles behind them, are organized and presented within a clear framework that allows for ease of comprehension. This text is intended for an advanced computer architecture course or a course in superscalar processor design. It is written at a level appropriate for senior or first year graduate level students, and can be used by professionals as well.

Performance Analysis and Verification of Super Scalar Processors

Performance Analysis and Verification of Super Scalar Processors
Author: International Business Machines Corporation. Research Division
Publisher:
Total Pages: 20
Release: 1995
Genre: Computer organization
ISBN:

Download Performance Analysis and Verification of Super Scalar Processors Book in PDF, Epub and Kindle

Abstract: "We discuss the problem of accurate analysis and tuning of pre-hardware processor performance. In particular, we focus our attention to a class of current high-end super scalar, RISC processors. We examine the spectrum of performance estimation and analysis tools from those used in early-stage design point definition to later-stage cycle- accurate simulation models. The problem of 'performance verification' is highlighted as one which is becoming increasingly important, as single- chip processor complexity increases to the point where 'designer intuition' alone is not enough to ensure elimination of major 'performance bugs.' We discuss currently used methods to validate performance estimates against expectations derived from the basic hardware and application constraints. Through judicious choice of methods or tools and corresponding validation aids, the designer is able to safeguard against introduction of performance inhibitors which may be hard to remedy after 'first silicon.' Also, using such analysis and verification techniques, the designer is able to prevent post-hardware surprises with respect to benchmark performance targets (e.g. SPECmarks). We illustrate our methodology in the context of realistic examples taken from the high-end PowerPC[superscript TM] processor arena."

Performance Study of Superthreaded Architectures

Performance Study of Superthreaded Architectures
Author: Zhenzhen Jiang
Publisher:
Total Pages: 160
Release: 1997
Genre: Computer architecture
ISBN:

Download Performance Study of Superthreaded Architectures Book in PDF, Epub and Kindle

Abstract: "This paper presents the simulation studies done on the Superthreaded Architecture. Two trace-driven, cycle-by-cycle Superthreaded processor simulators are implemented for the study. One issues single instruction per thread in each clock cycle, (called SIPT Superthreaded Simulator in this paper), the other has Superscalar features incorporated into it and issues multiple instructions per thread in each clock cycle (called MIPT Superthreaded Simulator in this paper). Both allow run-time data-dependence checking and instruction scheduling. The simulation results show that the Superthreaded architecture, which adopts a thread-pipelining execution model and allows threads with data dependencies and control dependencies to be executed in parallel, can handle loops with run-time control speculation very well, and can achieve good speedups for most of the SPEC benchmark programs. What's more, some design features of the existing single-threaded multiple-issue processor, such as Superscalar, can be added onto the Superthreaded architecture to further exploit instruction-level parallelism."