WORKLOAD CHARACTERIZATION FOR COMPUTER SYSTEM DESIGN

Similar documents
Perspectives of Information Systems

COMMUNICATION-BASED SYSTEMS

THE PROMOTION OF SOCIAL AWARENESS

Guide to Teaching Computer Science

Feature-oriented vs. Needs-oriented Product Access for Non-Expert Online Shoppers

PRODUCT PLATFORM AND PRODUCT FAMILY DESIGN

Computer Organization I (Tietokoneen toiminta)

PeopleSoft Human Capital Management 9.2 (through Update Image 23) Hardware and Software Requirements

Instrumentation, Control & Automation Staffing. Maintenance Benchmarking Study

International Series in Operations Research & Management Science

Education for an Information Age

Computer Architecture CSC

Automating Outcome Based Assessment

TIMSS ADVANCED 2015 USER GUIDE FOR THE INTERNATIONAL DATABASE. Pierre Foy

Managing Printing Services

Rule discovery in Web-based educational systems using Grammar-Based Genetic Programming

CPMT 1303 Introduction to Computer Technology COURSE SYLLABUS

DIGITAL GAMING & INTERACTIVE MEDIA BACHELOR S DEGREE. Junior Year. Summer (Bridge Quarter) Fall Winter Spring GAME Credits.

Intel-powered Classmate PC. SMART Response* Training Foils. Version 2.0

On-Line Data Analytics

MMOG Subscription Business Models: Table of Contents

Improving Memory Latency Aware Fetch Policies for SMT Processors

GACE Computer Science Assessment Test at a Glance

McGraw-Hill Connect and Create Built by Blackboard. Release Notes. Version 2.3 for Blackboard Learn 9.1

USER ADAPTATION IN E-LEARNING ENVIRONMENTS

Improving Fairness in Memory Scheduling

EDUCATION IN THE INDUSTRIALISED COUNTRIES

CWIS 23,3. Nikolaos Avouris Human Computer Interaction Group, University of Patras, Patras, Greece

Institutionen för datavetenskap. Hardware test equipment utilization measurement

Android App Development for Beginners

Measurement & Analysis in the Real World

Ph.D. Computer Engineering and Information Science. Case Western Reserve University. Cleveland, OH, 1986

Communication and Cybernetics 17

Implementing a tool to Support KAOS-Beta Process Model Using EPF

K 1 2 K 1 2. Iron Mountain Public Schools Standards (modified METS) Checklist by Grade Level Page 1 of 11

MARE Publication Series

Elearn: Towards a Collaborative Educational Virtual Environment

Ericsson Wallet Platform (EWP) 3.0 Training Programs. Catalog of Course Descriptions

Library Consortia: Advantages and Disadvantages

Modeling user preferences and norms in context-aware systems

Courses in English. Application Development Technology. Artificial Intelligence. 2017/18 Spring Semester. Database access

MINISTRY OF EDUCATION

CPMT 1347 Computer System Peripherals COURSE SYLLABUS

Python Machine Learning

Louisiana Free Materials List

Problem Solving for Success Handbook. Solve the Problem Sustain the Solution Celebrate Success

Evaluation of Usage Patterns for Web-based Educational Systems using Web Mining

Evaluation of Usage Patterns for Web-based Educational Systems using Web Mining

Course Specifications

Diploma in Library and Information Science (Part-Time) - SH220

1 Use complex features of a word processing application to a given brief. 2 Create a complex document. 3 Collaborate on a complex document.

Software Development Plan

Hongyan Ma. University of California, Los Angeles

IMPLEMENTING EUROPEAN UNION EDUCATION AND TRAINING POLICY

MASTER OF SCIENCE (M.S.) MAJOR IN COMPUTER SCIENCE

INFED. INFLIBNET Access Management Federation Yatrik Patel

FY16 UW-Parkside Institutional IT Plan Report

Chapter 7 Information and Communications Technology: Platforms for Learning and Teaching

A Practical Approach to Embedded Systems Engineering Workforce Development

Using Virtual Manipulatives to Support Teaching and Learning Mathematics

Worldwide Online Training for Coaches: the CTI Success Story

An Introduction to Simio for Beginners

Bluetooth mlearning Applications for the Classroom of the Future

Circuit Simulators: A Revolutionary E-Learning Platform

Online Marking of Essay-type Assignments

Running Head: STUDENT CENTRIC INTEGRATED TECHNOLOGY

A Cost-Effective Cloud Service for E-Learning Video on Demand

Ministry of Education and Science of Kazakhstan. Karaganda State Technical University

AUTONOMY. in the Law

Student Information System. Parent Quick Start Guide

Specification and Evaluation of Machine Translation Toy Systems - Criteria for laboratory assignments

Distributed Weather Net: Wireless Sensor Network Supported Inquiry-Based Learning

Business Analytics and Information Tech COURSE NUMBER: 33:136:494 COURSE TITLE: Data Mining and Business Intelligence

Instructor Experience and Qualifications Professor of Business at NDNU; Over twenty-five years of experience in teaching undergraduate students.

BUS Computer Concepts and Applications for Business Fall 2012

Computer Science. Embedded systems today. Microcontroller MCR

Lecture Notes in Artificial Intelligence 4343

SOFTWARE EVALUATION TOOL

IAT 888: Metacreation Machines endowed with creative behavior. Philippe Pasquier Office 565 (floor 14)

New Paths to Learning with Chromebooks

Speech Recognition at ICSI: Broadcast News and beyond

Education: Integrating Parallel and Distributed Computing in Computer Science Curricula

Focus on. Learning THE ACCREDITATION MANUAL 2013 WASC EDITION

An OO Framework for building Intelligence and Learning properties in Software Agents

Collaboration: Meeting the Library User's Needs in a Digital Environment

Lectora a Complete elearning Solution

Hardhatting in a Geo-World

PH.D. IN COMPUTER SCIENCE PROGRAM (POST M.S.)

Regan's Resume Last Edit : 31 March 2008

Blank Table Of Contents Template Interactive Notebook

Learning to Schedule Straight-Line Code

3. Improving Weather and Emergency Management Messaging: The Tulsa Weather Message Experiment. Arizona State University

Specification of the Verity Learning Companion and Self-Assessment Tool

HMS 241 Lab Introduction to Early Childhood Education Fall 2015

ebusiness Technologies Spring 2000 Syllabus

MAHATMA GANDHI KASHI VIDYAPITH Deptt. of Library and Information Science B.Lib. I.Sc. Syllabus

A Coding System for Dynamic Topic Analysis: A Computer-Mediated Discourse Analysis Technique

Designing Educational Computer Games to Enhance Teaching and Learning

similar to the majority ofcomputer science courses in colleges and universities today. Classroom time consisted of lectures, albeit, with considerable

Automating the E-learning Personalization

Transcription:

WORKLOAD CHARACTERIZATION FOR COMPUTER SYSTEM DESIGN

THE KLUWER INTERNATIONAL SERIES IN ENGINEERING AND COMPUTER SCIENCE

WORKLOAD CHARACTERIZATION FOR COMPUTER SYSTEM DESIGN edited by Lizy Kurian John The University of Texas at Austin Ann Marie Grizzaffi Maynard IBM Austin Research Laboratory SPRINGER SCIENCE+BUSINESS MEDIA, LLC

Library of Congress Cataloging-in-Publication Data Workload characterization for computer system design/edited by Lizy Kurian John, Ann Marie Grizzaffi Maynard. p.cffi. - (The Kluwer international series in engineering and computer science ; SECS 542) Includes bibliographical references and index. ISBN 978-1-4613-6973-8 ISBN 978-1-4615-4387-9 (ebook) DOI 10.1007/978-1-4615-4387-9 1. System design. 2. Computer systems. 1. John, Lizy Kurian. II. Maynard, Ann Marie Grizzaffi. ID. Series. QA76.9.S88 W68 2000 oo4.2'i-dc21 99.Q85691 Copyright 2000 by Springer Science+Business Media New York Originally published by Kluwer Academic Publishers, New York in 2000 Softcover reprint ofthe hardcover Ist edition 2000 AII rights reserved. No part of this publication may be reproduced, stored in a retrieval system or transmitted in any form or by any means, mechanical, photo-copying, recording, or otherwise, without the prior written permission of the publisher, Springer Science+Business Media, LLC. Printed on acid{ree paper.

Contents Preface... vii 1. UltraSparc Instruction Level Characterization of Java Virtual Machine Workload... 1 Andrea Barisone, Francesco Bellotti, Riccardo Berta and Alessandro De Gloria 2. Analyzing Memory Reference Traces of Java Programs... 25 Jin-Soo Kim and Yarsun Hsu 3. Towards a Simplified Database Workload for Computer Architecture Evaluations... 49 Kimberly Keeton and David A Patterson 4. Characterization of Bus Transactions for SPECweb96 Benchmark... 73 Prasant Mohapatra, Hariharan Thantry and Krishna Kant 5. Performance Impact of Un cached File Accesses in SPECweb99... 87 Krishna Kant and Youjip Won 6. Performance of Natural 110 Applications... 105 Stevan Vlaovic and Richard Uhlig 7. Workloads for Programmable Network Interfaces... 135 Patrick Crowley, Marc E. Fiuczynski, Jean-Loup Baer and Brian Bershad

vi 8. A Characterization of Control Independence in Programs... 149 Anasua Bhowmik and Mano} Franklin 9. Characterization of Graphics Activities in PC Benchmarks by Bottleneck Detection... 171 Daniel Etiemble 10. Workload ofa Media-Enhanced Classroom Server... 191 Nissim Harel, Vivekanand Vellanki, Ann Chervenak, Gregory Abowd and Umakishore Ramachandran Index... 211

Preface The advent of the world-wide web and web-based applications have dramatically changed the nature of computer applications. Computer system design, in the light of these changes, involves understanding these modem workloads, identifying bottlenecks during their execution, and appropriately tailoring microprocessors, memory systems, and the overall system to minimize bottlenecks. This book contains ten chapters dealing with several contemporary programming paradigms including Java, web server and database workloads. The first two chapters concentrate on Java. While Barisone et al.'s characterization in Chapter 1 deals with instruction set usage of Java applications, Kim et al.'s analysis in Chapter 2 focuses on memory referencing behavior of Java workloads. Several applications including the SPECjvm98 suite are studied using interpreter and Just-In-Time (TIT) compilers. Barisone et al.'s work includes an analytical model to compute the utilization of various functional units. Kim et al. present information on locality, live-range of objects, object lifetime distribution, etc. Studying database workloads has been a challenge to research groups, due to the difficulty in accessing standard benchmarks. Configuring hardware and software for database benchmarks such as those from the Transactions Processing Council (TPC) requires extensive effort. In Chapter 3, Keeton and Patterson present a simplified workload (microbenchmark) that approximates the characteristics of complex standardized benchmarks. Chapters 4 and 5 are about web server workloads. The Standard Performance Evaluation Co-operative (SPEC) has released two suites of web

V III benchmarks, the SPECweb96 that contains static web applications and the SPECweb99 that contains dynamic web server applications. In Chapter 4, Mohapatra et al. present characterization of bus traffic generated by the SPECweb96 suite and in Chapter 5, Kant et al. investigate the file caching properties of SPECweb99 benchmarks. Processing of speech, handwriting, gestures and other natural forms of input-output has become extremely important with the popularity of personal design assistants. Vlaovich and Uhlig study several natural 110 applications such as ParaScript LLC, Naturally Speaking, Power Translator, etc. in Chapter 6. A comparison of the features of these applications with SPEC95 is also presented. Network interface workloads performing network address translation, TCP connection management, packet filtering, packet forwarding, data encryption, etc. are studied by Crowley et al. in Chapter 7. The characteristics of network interface workloads are studied on popular execution paradigms such as superscalar and multithreaded architectures. The available parallelism in programs is constrained by control flow changes (branches), however several sections of the code are seen to be executed irrespective of the outcome of the control instructions. In Chapter 8, Bhowmik and Franklin investigate the control independence in programs and demonstrate that available parallelism in ordinary programs improve significantly with exploitation of control independence. Chapter 9 presents system level characterization of 20 and 3D graphics applications on desktop PCs, and identifies whether the CPU, memory, disk, or graphics board is the bottleneck in the system. Chapter 10 presents system level characterization such as server bandwidth, network bandwidth, and storage requirements of a server that deals with audio and video information in a media enhanced University classroom. While the characterization in many initial chapters is at a processor level, Chapters 9 and 10 present system level studies. The chapters in this book are revised versions of papers presented at the Second Annual Workshop on Workload Characterization (WWC'99) held in Austin, Texas, on October 9, 1999. First of all we would like to thank the authors of the various chapters for their contributions. We also wish to express our sincere gratitude to all those who reviewed manuscripts for the book. Juan Rubio of the Laboratory for Computer Architecture (LCA) at the University of Texas at Austin significantly helped in the creation of this

Preface ix book and deserves special acknowledgment. Many other LCA members also helped in the process. We wish to thank them all. The financial support provided by Intel Corporation is gratefully acknowledged. We also take this opportunity to thank all attendees of the workshop for the informative and enjoyable discussions. Lizy Kurian John, The University of Texas at Austin Ann Marie Grizzaffi Maynard, IBM Austin Research Laboratory