Skip to main content

Application and Program Performance Analysis Using PEX Statistics on IBM i5/OS

An IBM Redbooks publication

Note: This is publication is now archived. For reference only.


Published on 25 July 2007

  1. .PDF (6.7 MB)

Share this page:   

ISBN-10: 0738489093
ISBN-13: 9780738489094
IBM Form #: SG24-7457-00

Authors: Jim Cook, Rick Turner and Deborah Hatt

    menu icon


    This IBM Redbooks publication is intended for use by those generally familiar with most of the iSeries IBM-provided performance tools available through the i5/OS operating system’s commands and the additional cost Performance Tools for iSeries, 5722-PT1, licensed program.

    i5/OS comes with a detailed program level performance data collection capability called the Performance Explorer (PEX). i5/OS commands supporting the collection include Add PEX Definition, Start Performance Explorer, and End Performance Explorer. One of the Performance Explorer (PEX) collection options is called Statistics (*STATS), which collects the program level performance statistics, including CPU usage, disk I/O activity, and the occurrence of certain i5/OS and System i microcode level events. The Print PEX Report function of 5722-PT1 provides a basic view of this *STATS data.

    PEX Statistics provides a richer interface for collection and analysis of the *STATS performance data than is available through the i5/OS PEX command and the Print PEX Report output.

    Table of Contents

    Chapter 1. Introduction to PEX Statistics

    Chapter 2. Getting started: using Stats Flat

    Chapter 3. Analyzing PEX Stats Flat data

    Chapter 4. Collecting PEX Stats Hierarchical data

    Chapter 5. Analyzing PEX Stats Hierarchical data

    Chapter 6. Additional PEX Stats data analysis topics

    Chapter 7. Performance Analysis topics

    Appendix A. Building a PEX Definition for PEX Stats collection

    Appendix B. iDoctor and PEX Analyzer options

    Appendix C. Querying and graphing PEX Statistics data

    Appendix D. PEX Stats user-defined queries

    Appendix E. Program and MI Complex Instructions summary and Stats' analysis file descriptions


    Others who read this also read