Front cover Reporting with TPCTOOL Learn the reporting capabilities of TPCTOOL Create customized reports Evaluate report data Massimo Mastrorilli ibm.com/redbooks Redpaper International Technical Support Organization Reporting with TPCTOOL August 2007 REDP-4230-00 Note: Before using this information and the product it supports, read the information in “Notices” on page v. First Edition (August 2007) This edition applies to Version 3 of IBM TotalStorage Productivity Center (product number 5608-VC0). © Copyright International Business Machines Corporation 2007. All rights reserved. Note to U.S. Government Users Restricted Rights -- Use, duplication or disclosure restricted by GSA ADP Schedule Contract with IBM Corp. Contents Notices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .v Trademarks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . vi Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . vii The team that wrote this Redpaper . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . vii Become a published author . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . vii Comments welcome. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . vii Chapter 1. TPCTOOL - what is it and when you should use it . . . . . . . . . . . . . . . . . . . . 1 1.1 TPCTOOL functionality . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 1.2 What do you need to install and run TPCTOOL - part 1 of 2 . . . . . . . . . . . . . . . . . . . . . 3 1.3 What do you need to install and run TPCTOOL - part 2 of 2 . . . . . . . . . . . . . . . . . . . . . 4 1.4 Where to install TPCTOOL . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 1.5 TPCCLI.CONF . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 1.6 How to run a command . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 1.6.1 Single-shot mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 1.6.2 Interactive mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 1.6.3 Multiple / Script Command Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 1.6.4 Output syntax to a file . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 1.7 Basic syntax . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 1.8 When to use TPCTOOL instead of GUI to get reports . . . . . . . . . . . . . . . . . . . . . . . . . 10 1.9 Reports with CLI interface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14 Chapter 2. TPCTOOL reports and configuration data . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1 TPC Server tasks before creating reports . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2 GUID, WWN - how to get them . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.3 Multiple metrics - Tabular reports . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.4 Multiple metrics - Graphical reports . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.5 Multiple metrics - TPCTOOL . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.6 TPCTOOL - Programming technique . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.7 Commands to start . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.8 How to create a graph from a text file . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15 16 17 18 19 20 21 22 24 Chapter 3. Rules of thumb . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1 The customer wants to know... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.2 Response time . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.3 Assumptions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.4 Performance metrics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.5 Look at historical data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.6 Performance metric guidelines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.7 RAID level and RPM considerations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25 26 27 29 30 31 32 34 Chapter 4. Quick start for disk performance monitoring . . . . . . . . . . . . . . . . . . . . . . . 4.1 Throughput and response time metrics. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.2 How to evaluate response time . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.3 Additional metrics related to throughput . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.4 Backend and frontend metrics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.5 Backend response time. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.6 Historical performance charts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37 38 39 41 44 45 48 © Copyright IBM Corp. 2007. All rights reserved. iii 4.7 Port Data Rate and Port Response time . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51 Chapter 5. Sample TPCTOOL reports . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.1 Data rate vs. response time for disk storage subsystem. . . . . . . . . . . . . . . . . . . . . . . . 5.2 Ports report for a disk storage subsystem. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.3 SVC performance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.4 Reports for a switch fabric. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53 54 56 58 61 Chapter 6. Macro to create charts from TPCTOOL CLI text file . . . . . . . . . . . . . . . . . . 6.1 Importing and exporting data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.1.1 TimeStamp . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.1.2 Create macros . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.1.3 Creating a template. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.1.4 Creating graphs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65 66 71 73 77 79 Chapter 7. Metrics per subsystem. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85 7.1 Metrics for DS4000 storage subsystem . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86 7.2 Metrics for ESS storage subsystem . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87 7.3 Metrics for DS8000/DS6000 storage subsystems . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95 7.4 Metrics for SVC storage subsystems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103 7.5 Metrics for switch fabric. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 116 Related publications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . IBM Redbooks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . How to get IBM Redbooks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Help from IBM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119 119 119 119 Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121 iv Reporting with TPCTOOL Notices This information was developed for products and services offered in the U.S.A. IBM may not offer the products, services, or features discussed in this document in other countries. Consult your local IBM representative for information on the products and services currently available in your area. Any reference to an IBM product, program, or service is not intended to state or imply that only that IBM product, program, or service may be used. Any functionally equivalent product, program, or service that does not infringe any IBM intellectual property right may be used instead. However, it is the user's responsibility to evaluate and verify the operation of any non-IBM product, program, or service. IBM may have patents or pending patent applications covering subject matter described in this document. The furnishing of this document does not give you any license to these patents. You can send license inquiries, in writing, to: IBM Director of Licensing, IBM Corporation, North Castle Drive, Armonk, NY 10504-1785 U.S.A. The following paragraph does not apply to the United Kingdom or any other country where such provisions are inconsistent with local law: INTERNATIONAL BUSINESS MACHINES CORPORATION PROVIDES THIS PUBLICATION "AS IS" WITHOUT WARRANTY OF ANY KIND, EITHER EXPRESS OR IMPLIED, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF NON-INFRINGEMENT, MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE. Some states do not allow disclaimer of express or implied warranties in certain transactions, therefore, this statement may not apply to you. This information could include technical inaccuracies or typographical errors. Changes are periodically made to the information herein; these changes will be incorporated in new editions of the publication. IBM may make improvements and/or changes in the product(s) and/or the program(s) described in this publication at any time without notice. Any references in this information to non-IBM Web sites are provided for convenience only and do not in any manner serve as an endorsement of those Web sites. The materials at those Web sites are not part of the materials for this IBM product and use of those Web sites is at your own risk. IBM may use or distribute any of the information you supply in any way it believes appropriate without incurring any obligation to you. Information concerning non-IBM products was obtained from the suppliers of those products, their published announcements or other publicly available sources. IBM has not tested those products and cannot confirm the accuracy of performance, compatibility or any other claims related to non-IBM products. Questions on the capabilities of non-IBM products should be addressed to the suppliers of those products. This information contains examples of data and reports used in daily business operations. To illustrate them as completely as possible, the examples include the names of individuals, companies, brands, and products. All of these names are fictitious and any similarity to the names and addresses used by an actual business enterprise is entirely coincidental. COPYRIGHT LICENSE: This information contains sample application programs in source language, which illustrate programming techniques on various operating platforms. You may copy, modify, and distribute these sample programs in any form without payment to IBM, for the purposes of developing, using, marketing or distributing application programs conforming to the application programming interface for the operating platform for which the sample programs are written. These examples have not been thoroughly tested under all conditions. IBM, therefore, cannot guarantee or imply reliability, serviceability, or function of these programs. © Copyright IBM Corp. 2007. All rights reserved. v Trademarks The following terms are trademarks of the International Business Machines Corporation in the United States, other countries, or both: AIX® DS4000™ DS6000™ DS8000™ IBM® Redbooks® Redbooks (logo) System Storage™ Tivoli® TotalStorage® ® The following terms are trademarks of other companies: Java, and all Java-based trademarks are trademarks of Sun Microsystems, Inc. in the United States, other countries, or both. Excel, Microsoft, Visual Basic, Windows, and the Windows logo are trademarks of Microsoft Corporation in the United States, other countries, or both. UNIX is a registered trademark of The Open Group in the United States and other countries. Linux is a trademark of Linus Torvalds in the United States, other countries, or both. Other company, product, or service names may be trademarks or service marks of others. vi Reporting with TPCTOOL Preface Since the introduction of TotalStorage® Multiple Device Manager, and its subsequent replacement by versions of TotalStorage Productivity Center, and ever since the stabilization of the TotalStorage ESS Expert, customers have asked for a way to pull performance metrics from the TotalStorage Productivity Center database in much the same way they could pull the metrics from the ESS Expert database. There are always a few leading edge customers who know what they want to do with performance monitoring and performance management, and they know how they want to do it. This IBM® Redpaper gives you an overview of the function of TPCTOOL and shows you how to use it to generate reports based on your TotalStorage Productivity Center repository data. The team that wrote this Redpaper This Redpaper was produced by a specialist from Switzerland working at the International Technical Support Organization, San Jose Center. Massimo Mastrorilli is an Advisory IT Storage Specialist in Switzerland. He joined IBM Italy in 1989 and seven years ago he moved to IBM Switzerland, based in Lugano. He has 16 years of experience in implementing, designing, and supporting Storage solutions in S390 and Open Systems environment. His area of expertise include IBM Tivoli® Storage Manager, SAN Storage Area Network and Storage solutions for Open Systems. He is an IBM Certified Specialist for TSM, Storage Sales and Open System Storage™ Solutions. He is a member of Tivoli GRT Global Response Team group. Thanks to Brian Smith for allowing us to convert his TPCTOOL presentation materials into this Redpaper and his continued support of this project. Become a published author Join us for a two- to six-week residency program! Help write an IBM Redbooks publication dealing with specific products or solutions, while getting hands-on experience with leading-edge technologies. You'll have the opportunity to team with IBM technical professionals, Business Partners, and Clients. Your efforts will help increase product acceptance and customer satisfaction. As a bonus, you will develop a network of contacts in IBM development labs, and increase your productivity and marketability. Find out more about the residency program, browse the residency index, and apply online at: ibm.com/redbooks/residencies.html Comments welcome Your comments are important to us! We want our papers to be as helpful as possible. Send us your comments about this Redpaper or other Redbooks® in one of the following ways: © Copyright IBM Corp. 2007. All rights reserved. vii Use the online Contact us review Redbooks form found at: ibm.com/redbooks Send your comments in an email to: redbooks@us.ibm.com Mail your comments to: IBM Corporation, International Technical Support Organization Dept. HYTD Mail Station P099 2455 South Road Poughkeepsie, NY 12601-5400 viii Reporting with TPCTOOL 1 Chapter 1. TPCTOOL - what is it and when you should use it This chapter provides an overview of TPCTOOL. It covers: The main function provided by TPCTOOL How to use TPCTOOL Where to find the syntax How to install TPCTOOL We also give an understanding of why you would use TPCTOOL and when to use it instead of the GUI. © Copyright IBM Corp. 2007. All rights reserved. 1 1.1 TPCTOOL functionality Figure 1-1 TPCTOOL Functionality There have been many requests for the ability to produce performance reports from the TPC database and produce multiple metric graphs as provided by other IBM Storage products. TPCTOOL is a command line (CLI) based program which interacts with the TPC Device server (see Figure 1-1). It allows you to create graphs and charts with multiple metrics, with different unit types and for multiple entities (for example, Subsystems, Volumes, Controller, Arrays). Commands are entered as lines of text (that is, sequences of types of characters) and output can be received as text. The command can be used to access Generic Commands, Device Server, Server Administration Configuration and Reporting. The example in Figure 1-1 shows the lsdev command, which lists all the devices discovered by the TPC server you are querying. The tool provides queries, management and reporting capabilities. You cannot initiate discoveries, probes and performance collection from the tool. With the release of TPC V3.1 Storage Resource Management Command Processor (SCRMCP), perfcli (Performance Management), and AgentCLI have been integrated into TPCTOOL. It is installable anywhere, the CLI code can be deployed on any computer that has access to the network where the TPC server is located. Connects via TCP/HTTP/SOAP to the Web Service API. Used for Storage provisioning and management. Standalone Java™ Client. 2 Reporting with TPCTOOL 1.2 What do you need to install and run TPCTOOL - part 1 of 2 Figure 1-2 How to install CLI to use TPCTOOL Before starting the installation, these are prerequisites that you should know or prepare: 1. You will need to have the TPC server code available to install the TPCTOOL CLI. – Start the TotalStorage Productivity Installer, select a Custom installation, and select the CLI as shown in Figure 1-2. – You need only Disk1 of the TPC code to install the CLI software. Chapter 1. TPCTOOL - what is it and when you should use it 3 1.3 What do you need to install and run TPCTOOL - part 2 of 2 Figure 1-3 What you need to install - part 2 of 2 1. You need IP connection to TPC Device Server. 2. You need the IP address and the port (the default is 9550) used for TPC Device Server. 3. You need to know the Host Authentication Password. This is specified during the installation of the Agent Manager. The default password is changeMe. 4. You may add to the TPCCLI.CONF file the user and password of a fabric or disk administrator (TPC Role based authority) to issue these commands. It is not mandatory that you create TPCCLI.CONF, TPCTOOL CLI works even without any TPCCLI.CONF file configured. But if you did not have the TPCCLI.CONF file configured then you have to specify a valid user and the password in all your commands. 4 Reporting with TPCTOOL 1.4 Where to install TPCTOOL Figure 1-4 Install TPCTOOL on your workstation According to the prerequisites in 1.2, “What do you need to install and run TPCTOOL - part 1 of 2” on page 3, you could install the TPCTOOL CLI on your computer (mobile computer, laptop) or on your workstation (see Figure 1-4). Table 1-1 shows the supported platforms for the TPCTOOL CLI. Table 1-1 Supported platforms Operating System Mode AIX® 5.3 32-bit and 64-bit (compatibility mode) Windows® 2000 Advanced Server 32-bit Windows 2000 Data Center 32-bit Windows 2003 Standard Edition 32-bit and 64-bit (compatibility mode) Windows 2003 Enterprise Edition 32-bit and 64-bit (compatibility mode) Windows XP 32-bit Red Hat Enterprise Linux® AS 3.0 32-bit IA32-xSeries Chapter 1. TPCTOOL - what is it and when you should use it 5 1.5 TPCCLI.CONF TPCCLI.conf sample acmeurl=10.8.66.77:9550 asvc1=10.8.140.146:0000020060C03052+0 asvc2=10.8.140.148:0000020060403200+0 Subsys1:=10.5.3.88:000000200656c9990+0 allsvcvolmetrics = 803,806,809,812,815,818,819,820,821,825,826,827 switchportmetrics= 855,856,857,858,859,860,861,862,869,870,871,872,873,874,875,876,877,878,879,880,88 1,882,883 myid=acmesuperuser mypwd=acmesuper asvc1vdisk=getrpt -url acmeurl -user myid -pwd mypwd -ctype 12 -columns 803,806,809,812,815,818,819,820,821,825,826,827,830,831,832,833 -level sample subsys asvc1 -fs ; asvc2vdisk=getrpt -url acmeurl -user myid -pwd mypwd -ctype 12 -columns 803,806,809,812,815,818,819,820,821,825,826,827,830,831,832,833 -level sample subsys asvc2 -fs Figure 1-5 tpccli.conf file Configuration files, or config files are used to configure the initial settings for computer operations. TPCTOOL allows you to you to utilize a config file to enter your settings to access your TPC device server and to execute syntax. Command is a textual substitution of the command string with a defined alias Parameters can be provided with default values that can be substituted Command and Parameter can be ‘aliased’ and stored in a config file on the client machine or server. The Aliased commands are saved in the command configuration file. TPCCLI.CONF file (see Figure 1-5) is a method to define alias commands that can be executed either in a script or though the interactive CLI interface. Note: The TPCCLI.CONF file is not provided by default The config file should be created and saved in the following directories: Windows: C:\Program Files\IBM\TPC\cli\libs Unix or Linux: /<usr or opt>/IBM/TPC/cli/libs The aliased commands are written in ASCII and are line oriented. The lines are terminated by a newline. The config file should be created by TPC administrators or SAN administrators. This config file needs to maintained on a regular basis depend on the changing environment and the function that is required. 6 Reporting with TPCTOOL 1.6 How to run a command How to run a command ode m tive terac d in le an Sing de t mo Scrip e ect th Redir a ut to outp file Figure 1-6 Command mode Since the CLI is divided into many different modes (see Figure 1-6), the commands available to you at any given time depend on the mode you are currently in. Entering a question mark (?) or help at the CLI prompt allows you to obtain a list of commands available for each command mode. 1.6.1 Single-shot mode Use the TPCTOOL CLI single-shot command mode if you want to issue a single occasional command. You must supply the login information and issue the command that you want to process at the same time. Perform the following steps to use the single-shot mode: 1. Start a Command WIndow (CMD) 2. You must set your path or CHDIR to <install>/cli. 3. From the <install>/cli directory, type your command at the shell prompt. Syntax: tpctool lsdev -user tpcadmin -pwd tpcadmin -url 9.43.85.143:9550 1.6.2 Interactive mode Use the TPCTOOL CLI interactive command mode when you have multiple transactions to process that cannot be incorporated into a script. The interactive command mode provides a history function that makes repeating or checking prior command usage easy to do. You may Chapter 1. TPCTOOL - what is it and when you should use it 7 enter the interactive mode by issuing the TPCTOOL command with no command line options. Perform the following steps to use the single-shot mode: 1. Start a Command WIndow (CMD). 2. You must set your path or CHDIR to <install>/cli. 3. From the <install>/cli directory, type tpctool at the shell prompt, you will now be within the interactive session. 4. At the prompt you may enter any valid TPCTOOL CLI command. Syntax: shell> tpctool tpctool> lsdev -user tpcadmin -pwd tpcadmin -url 9.43.85.143:9550 1.6.3 Multiple / Script Command Mode You can create a file that contains multiple TPCTOOL CLI commands (“TPCCLI.CONF” on page 6). Login commands can be included in the command file. Use the TPCTOOL CLI Multiple / Script command mode if you want to issue a sequence of CLI commands. Administrators can use this mode to create automated processes; for example, establishing a volume performance report for SVC. Consider the following when using the TPCTOOL CLI Multiple / Script command mode: The TPCTOOL CLI script can contain only TPCTOOL CLI commands. Use of shell commands results in a process failure. You can add comments to the scripts. Comments must be prefixed by the number sign (#); for example, # This script contains a list of metrics available for DS8000™ subsystem volume performance. Syntax: shell> tpctool -script 1.6.4 Output syntax to a file For programs that display a lot of text like TPCTOOL, consider redirecting text that is usually displayed to a file. Displaying a lot of text will slow down execution; scrolling text in a terminal windows on a workstation can cause an I/O bottleneck and use more CPU time. The command in Example 1-1 shows how to run the tool more efficiently by redirecting output to a file and then displaying the program output. Example 1-1 Redirecting output to a file tpctool lsdev -user ***** -pwd ***** -url localhost:9550 > C:\reports\Output.txt The file could be imported into Excel® or accessed to create custom reports. 8 Reporting with TPCTOOL 1.7 Basic syntax H elp for com m and sintax rf a c In te 0. L in e 7 7- 0 and 2 -1 7 om m 3 C C to r ,G R e fe n c e v 3 .1 re R e fe Figure 1-7 Help For a complete description of all commands to use with TPCTOOL, refer to the manual: IBM TotalStorage Productivity Center Command Line Interface Reference, GC32-1777 Type HELP to get the list of all available commands (see Figure 1-7). TPCTOOL has a help function to assist you with the syntax of each specific command. Enter - help | h | -? to get the details for each command. In this document, we show examples for the following commands: lsdev lsmetrics lstype getrpt Chapter 1. TPCTOOL - what is it and when you should use it 9 1.8 When to use TPCTOOL instead of GUI to get reports Prior to TPC V3.1.3 the GUI interface methods (Query or Batch) have a limitation in that they restrict the user to a presentation of a single performance metric per report. With TPC V3.1.3, you can now select multiple metrics so that you can compare metrics side by side in the same view. Only metrics of the same unit type can be displayed in a chart. A performance metric is represented by a single variable such as the TOTAL Input/Output operations performed per second as seen at the I/O group (Total IOPS). While this is good information it is only one value of many that must be understood when analyzing a performance problem. Typical metrics that are reviewed are IOPS for read and write requests, read and write data rates, and read and write response times. All of these metrics are needed and defined data points along the entire flow of data from the client host, through the SAN, SVC, and finally to the lower storage subsystem. While this method can provide valuable data for performance analyzing as it provides both details and help on how to interpret the data. Figure 1-8 shows how to select multiple metrics. If you choose metrics with different unit type, you get an error as shown in Figure 1-9. Figure 1-8 Selecting multiple metrics, different unit type. 10 Reporting with TPCTOOL Figure 1-9 Only metrics with the same unit type can be selected With TPC V3.1.3 you can select multiple metrics with same unit type and produce graphs, as shown in Figure 1-10. Chapter 1. TPCTOOL - what is it and when you should use it 11 Figure 1-10 Selecting different metrics, but with same unit type Then, using the GUI, you get a chart like the one in Figure 1-11. 12 Reporting with TPCTOOL Figure 1-11 Chart with multiple metrics from GUI, same unit type Important: The chart above can be produced only since TPC V3.1.3 and follow-on versions. Chapter 1. TPCTOOL - what is it and when you should use it 13 1.9 Reports with CLI interface Figure 1-12 Data Rate versus Response Time The CLI interface methods have an attractive alternative. The CLI method supports the capability either through the interactive or script interface to collect the data from the TPC data repository by start time and duration, but also for multiple metrics on a single report. You can do this even selecting metrics with different unit types. Figure 1-12 shows two different metrics in the same chart: Total Data Rate Overall Response Time The definition of the performance output should include best practice efforts by the client and storage product teams to provide the most appropriate result. 14 Reporting with TPCTOOL 2 Chapter 2. TPCTOOL reports and configuration data This chapter describes which information you can get using TPCTOOL CLI. It contains basic information about how to generate reports and obtain configuration data. Performance report examples are covered in this chapter as well as prerequisites to be checked before using TPCTOOL. © Copyright IBM Corp. 2007. All rights reserved. 15 2.1 TPC Server tasks before creating reports central TPC DB PING PROBE DISCOVERY Fabric Agent PERFORMANCE MONITOR SCAN Computers Data Agent SAN Components SNMP / API Storage and Tape Subsystems CIMOM Data Collection Jobs Figure 2-1 Prerequisites for performance report generation Before creating performance reports using the TPCTOOL CLI you have to successfully complete these steps on your TPC Server (see Figure 2-1): Successfully discover and configure CIMOMs for the storage subsystems on which you want to create reports. – Administrative services → Discovery → CIMOM Successfully Probed these storage subsystems; – TotalStorage Productivity Center → Monitoring → Probes Successfully run a Performance Monitor jobs against them. – Disk Manager → Monitoring → Subsystem Performance Monitors Similar techniques under Fabric Manager in the TPC Navigation Tree are used to collect performance data from switches. 16 Reporting with TPCTOOL 2.2 GUID, WWN - how to get them Figure 2-2 lsdev -perf -l Most TPCTOOL commands requires you specify the globally-unique identifier (GUID) for storage subsystems or WWN for switches and fabrics. Use the lsdev command to get information about storage subsystems, fabrics, and switches. This information includes the GUID, user-defined name, device type, status, and the time that the status was updated. You must have Disk administrator authority to use this command. The commands in Figure 2-2 help you collect this information to be used during your commands and scripts. These are the options used: -user Specifies a valid TotalStorage Productivity Center user ID. The user variable is a valid user ID. -pwd Specifies the password for the TotalStorage Productivity Center user ID. The password variable is the password. -url Specifies the Device server. The format of the URL variable is system:port_number, where system is either the host name or IP address, and port_number is a valid port number for the HTTP service of the Device server. The default port is 9550. -perf Specifies that the devices for which performance data is collected should be listed. You must have the applicable authority to view the devices. -l Specifies that the long version should be listed: – – – – – GUID User-defined name Device type Status Time that the status was update. If you omit this parameter, only the GUID is listed. Important: Note that all GUID and WWN are case sensitive! Chapter 2. TPCTOOL reports and configuration data 17 2.3 Multiple metrics - Tabular reports Figure 2-3 Tabular reports Using the TPC GUI, you can get Tabular reports (see Figure 2-3) with: Single time sample Multiple performance metrics Select: Disk Manager → Reporting → Storage Subsystem Performance → By Volume 18 Reporting with TPCTOOL 2.4 Multiple metrics - Graphical reports Figure 2-4 Graphical reports Graphical reports (see Figure 2-4) give: Multiple Time Samples But only one metric Select the following path from the TPC Navigation Tree: Disk Manager → Reporting → Storage Subsystem Performance → By Volume In the report pane click Generate Report and select additional volumes. Chapter 2. TPCTOOL reports and configuration data 19 2.5 Multiple metrics - TPCTOOL Figure 2-5 Multiple metrics - TPCTOOL There is a requirement to get Multiple metrics in multiple intervals, using different unit types, in addition, to analyze the data using SAS or MS Excel. Using the TPCTOOL CLI you can generate tabular reports with multiple metrics, multiple samples and different unit types including some or all metrics (see Figure 2-5). Important: With TPC V3.1.3, you can produce charts and graphs with multiple metrics using the TPC GUI. However, you cannot select metrics with different unit types, as you can do with TPCTOOL. To create this chart, we exported the text file produced by the tpctool getrpt command to Excel spreadsheet and we customized the chart. Details about how to create this are covered in 2.7, “Commands to start” on page 22. 20 Reporting with TPCTOOL 2.6 TPCTOOL - Programming technique Figure 2-6 Programming technique With aliasing, you define a name for the alias followed by a value that is the name of a command and any options associated with command. The password is automatically encrypted using the same encryption algorithm as the password command before being stored in the config file. In conjunction with the interactive mode, this enables secure password encryption (plain text passwords will not appear in a command line). Aliased commands are saved in the command configuration file. For additional information you should refer to the book IBM TotalStorage Productivity Center Command Line Interface Reference, GC32-1777. Using the output created by TPCTOOL, you can create custom reports, and personalize them to your needs. You can schedule TPCTOOL commands using cron jobs (UNIX® platform) or as scheduled Windows tasks. You cannot create a Constraint Violations report using TPCTOOL CLI. You must use TPC GUI for this task following the Navigation Tree path: Disk Manager → Reporting → Storage Subsystem Performance → Constraint violations TPCTOOL can be used for daily checks against your environment and to run your first level analysis. As soon as you get the initial information, you can drill down using the TPC GUI. Refer to Chapter 8 in the book IBM TotalStorage Productivity Center: The Next Generation, SG24-7194, for more information. Chapter 2. TPCTOOL reports and configuration data 21 2.7 Commands to start Figure 2-7 Commands to start Starting with the lsdev command mentioned in 2.2, “GUID, WWN - how to get them” on page 17, you can run: lstype to list the components known to TPC lsmetrics to list the performance metrics available for a component lscomp to list the components for which performance metrics have actually been collected (Performance Monitors started): lstime to list times for which performance metrics exist. For all these commands (except for lstype), you have to specify: – -user Specifies a valid TotalStorage Productivity Center user ID. The user variable is a valid user ID. – -pwd Specifies the password for the TotalStorage Productivity Center user ID. The password variable is the password. – -url Specifies the Device server. This is the format of the URL variable: system:port_number, where system is either the host name or IP address, and port_number is a valid port number for the HTTP service of the Device server (Default is 9550). Following is the syntax used in the example of Figure 2-7: tpctool lstype 22 Reporting with TPCTOOL tpctool lsmetrics -user tpcadmin -pwd tpcadmin -url 9.43.85.142:9550 -ctype 10 -subsys 2105.22513+0 Where: – -ctype specifies that the output should include only components of the specified type. See the lstype command for more information about the comp_type variable. – -subsys specifies the storage subsystem. The subsystem variable is the GUID of the storage subsystem. You can use the lsdev command to return information, including the GUID, for all storage subsystems that are discovered by TotalStorage Productivity Center. Important: From the output of lsmetrics command, you get the numeric value for each Metric. This has to be specified when you create report using tpctool getrpt as described in Chapter 5, “Sample TPCTOOL reports” on page 53. The example in Figure 2-7 shows metrics for ESS subsystem at Array level (ctype 10): 821 - Total Data Rate 822 - Read Response time tpctool lscomp -user tpcadmin -pwd tpcadmin -url 9.43.85.142:9550 -ctype switch_port -level sample -fabric 1000000051E34E895 -start 2006.10.10:09:00:00 -duration 86400 Where: – -fabric GUID specifies the fabric. The GUID variable is the globally unique identifier. – -subsys subsystem specifies the storage subsystem. The subsystem variable is the GUID of the storage subsystem. You can use the lsdev command to return information, including the GUID, for all storage subsystems that are discovered by TotalStorage Productivity Center. – -level sample | hourly | daily specifies the level for which the performance metrics of the components should be summarized. You can specify a sample summary, an hourly summary, or a daily summary. – -ctype comp_type specifies that the output should include only components of the specified type. See the lstype command for more information about the comp_type variable. – -start date/time specifies the date and time to start the sampling period. The date and time are formatted as: yyyy.MM dd:HH:mm:ss All time zones are relative to the Device server. See the lstime command for more information. – -duration duration_seconds specifies the duration of the sampling period, in seconds. The duration_seconds variable is an integer. tpctool lstime -user tpcadmin -pwd tpcadmin -url 9.43.85.142:9550 -ctype switch_port -level -sample -fabric 1000000051E34E895 Where: – -fabric GUID specifies the fabric. The GUID variable is the globally unique identifier. – -level sample | hourly | daily specifies the level for which the performance metrics of the components should be summarized. You can specify a sample summary, an hourly summary, or a daily summary. – -ctype comp_type specifies that the output should include only components of the specified type. See the lstype command for more information about the comp_type variable. Chapter 2. TPCTOOL reports and configuration data 23 2.8 How to create a graph from a text file Figure 2-8 How to create a graph from a text file You can direct the output of TPCTOOL CLI to a text file using the command: tpctool lstype > output.txt Then, you can import this text file into an Excel spreadsheet or similar tool. Refer to Chapter 6, “Macro to create charts from TPCTOOL CLI text file” on page 65 for detailed information about this step. 24 Reporting with TPCTOOL 3 Chapter 3. Rules of thumb This chapter describes which information you can get using TPCTOOL CLI. It contains basic information about how to generate reports and obtain configuration data. Performance report examples and prerequisites to be checked before using TPCTOOL are covered in this chapter as well. © Copyright IBM Corp. 2007. All rights reserved. 25 3.1 The customer wants to know... Figure 3-1 Rules of thumb Customers want to know typical values for their performance metrics – rules of thumb or best practices. It is truly difficult to provide a simple answer for this question. The throughput for storage volumes can range from fairly small numbers (1 to 10 I/O per second) to very large values (more than 1000 I/O per second). This depends on the nature of the application. Note: When the I/O rates (throughput) approach 1000 I/O per second per volume, it is because the volume is getting good performance, usually from good cache behavior. 26 Reporting with TPCTOOL 3.2 Response time Figure 3-2 Response time We often assume (and our performance models assume) that 10 milliseconds (ms) is fairly high. But for a particular application, 10 ms may be too low or too high. Many On-Line Transaction Processing (OLTP) environments require response times closer to 5 ms, while batch applications with large sequential transfers may be fine with 20 ms response time. The appropriate value may also change between shifts or on the weekend. A response time of 5 ms may be required from 8 until 5, while 50 ms is perfectly acceptable near midnight. It is all customer and application dependent. The value of 10 ms is somewhat arbitrary, but related to the nominal service time of current generation disk products. In general terms, the service time of a disk is composed of a seek, a latency, and a data transfer. Nominal seek times these days can range from 4 to 8 ms, though in practice, many workloads do better than nominal. It is not uncommon for applications to experience from 1/3 to 1/2 the nominal seek time. Latency is assumed to be 1/2 the rotation time for the disk, and transfer time for typical applications is less than a millisecond. Note: So it is not unreasonable to expect 5-7 ms service time for a simple disk access. Under ordinary queuing assumptions, a disk operating at 50% utilization would have a wait time roughly equal to the service time. So 10-14 ms response time for a disk is not unusual, and represents a reasonable goal for many applications. For cached storage subsystems, we certainly expect to do as well or better than uncached disks, though that may be harder than you think. If there are a lot of cache hits, the subsystem Chapter 3. Rules of thumb 27 response time might be well below 5 ms, but poor read hit ratios and busy disk arrays behind the cache will drive the average response time number up. A high cache hit ratio allows us to run the backend storage ranks at higher utilizations than we might otherwise be satisfied with. Rather than 50% utilization of disks, we might push the disks in the ranks to 70% utilization, which would produce high rank response times, which are averaged with the cache hits to produce acceptable average response times. Conversely, poor cache hit ratios require pretty good response times from the backend disk ranks in order to produce an acceptable overall average response time. 28 Reporting with TPCTOOL 3.3 Assumptions Figure 3-3 Assumptions To make a long story short, front end response times probably need to be in the 5-15 ms range. The rank (backend) response times can usually operate in the 20-25 ms range unless the hit ratio is really poor. Backend write response times can be even higher, generally up to 80 ms (see Figure 3-3). There are applications (typically batch applications) for which response time is not the appropriate performance metric. In these cases, it is often the throughput in megabytes per second that is most important, and maximizing this metric will drive response times much higher than 30 ms. But this is a quick start document, and will not deal extensively with such batch workloads. Chapter 3. Rules of thumb 29 3.4 Performance metrics Figure 3-4 Performance metrics The most important metrics are throughput and response time metrics. These reports are available for different storage components. Figure 3-4 shows the different component level for which you can produce reports. Not all subsystems provide the same level of detail, but that will gradually change over time, as SMI-S standards evolve. In addition to throughput graphs, you may also produce graphs (or tabular data) for any of the metrics listed in Chapter 7, “Metrics per subsystem” on page 85. For example, if the Write Response Time becomes high, you might want to look at the NVS Full metric for various components, such as the Volume or Disk Array. The Read, Write, and Overall Transfer Sizes are useful for understanding throughput and response times. 30 Reporting with TPCTOOL 3.5 Look at historical data Figure 3-5 Look at historical data For most components, whether box, cluster, array or port, there will be expected limits to many of the performance metrics. But there are few rules of thumb, because it depends so much on the nature of the workload. Online Transaction Processing (OLTP) is so different from Backup (for instance, TSM Backup) that the expectations cannot be similar. OLTP is characterized by small transfers, and consequently data rates may be lower than the capability of the array or box hosting the data. TSM Backup uses large transfer sizes, so the I/O rates may seem low, yet the data rates test the limits of individual arrays (RAID ranks). And each box has different performance characteristics, from ESS F20, ESS 800 (including Turbo), SVC, DS4000™, DS6000™, to DS8000 models (Figure 3-5), each box will have different expectations for each component. The best rules of thumb are derived from looking at current (and historical) data for the configuration and workloads that are not getting complaints from their users. From this performance base, you can do trending, and in the event of performance complaints look for the changes in workload that can cause them. Chapter 3. Rules of thumb 31 3.6 Performance metric guidelines Figure 3-6 Performance metric guidelines Keeping in mind considerations made in the 3.5, “Look at historical data” on page 31 above, here are some metrics and limits that usually make sense. At least, these will provide a starting comparison, to see how your particular environment compares to these numbers, and then to understand why (Figure 3-6). Small block reads (4-8KB/Op) should have average response times in the 2 ms to 15 ms range. The low end of the range comes from very good Read Hit Ratio, while the high end of the range may represent either lower hit ratio or higher I/O rates. Average response times can also vary from time interval to time interval. It is not uncommon to see some intervals with higher response times, Small block writes should have response times near 1 ms. These should all be writes to cache and NVS and be very fast, unless the write rate exceeds the NVS and rank capabilities. Performance metrics for these considerations will be discussed later. Large reads (32 KB or greater) and large writes often signify batch workloads or highly sequential access patterns. These environments often prefer high throughput to low response times, so there is no guideline for these I/O characteristics. Batch and overnight workloads can tolerate very high response times without indicating problems. Read Hit Percentages can vary from near 0% to near 100%. Anything below 50% is considered low, but many database applications show hit ratios below 30%. For very low hit ratios, you need many ranks providing good backend response time. It is difficult to predict whether more cache will improve the hit ratio for a particular application. Hit ratios are more dependent on the application design and amount of data than on the size of cache (especially for Open System workloads). But larger caches are always better than 32 Reporting with TPCTOOL smaller ones. For high hit ratios, the backend ranks can be driven a little harder, to higher utilizations. For Random Read I/O, the backend rank (disk) read response times should seldom exceed 25 ms, unless the read hit ratio is near 99%. Backend Write Response Times will be higher because of RAID 5 (or RAID 10) algorithms, but should seldom exceed 80 ms. There will be some time intervals when response times exceed these guidelines. Chapter 3. Rules of thumb 33 3.7 RAID level and RPM considerations Figure 3-7 RAID and RPM RAID ranks have I/O per second limitations that depend on the type of RAID (RAID5 versus RAID10) and the number of disks in the rank. Because of the different RAID algorithms, it is not easy to know how many I/Os are actually going to the backend RAID ranks. For many RAID 5 subsystems, a worst case scenario can be approximated by using the backend read rate plus four times the backend write rate (R + 4 * W) where R and W are the backend read rate and backend write rate. Sequential writes can behave considerably better than worst case. Use care when trying to estimate the number of backend Ops to a RAID rank. The performance metrics seldom report this number precisely. You have to use the number of backend read and write operations to deduce an approximate backend Ops/sec number. The rank I/O limit depends on many factors, chief among them are the number of disks in the rank and the speed (RPM) of the disks. Note: When the number of I/O per second to a rank is near or above 1000, the rank should be considered very busy! For 15K RPM disks, the limit is a bit higher. But these high I/O rates to the backend ranks are not consistent with good performance; they imply the backend ranks are operating at very high utilizations, indicative of considerable queuing delays. Good capacity planning demands a solution that reduces the load on such busy ranks. Let us consider the upper limit of performance for 10K and 15K RPM, enterprise class devices. Be aware that different people have different opinions about these limits, but rest assured that all these numbers (see Figure 3-7) represent very busy DDMs. 34 Reporting with TPCTOOL DDM Speed Max Ops/sec 6+P Ops/sec 7+P Ops/sec 10K RPM 150-175 1050-1225 1200-1400 15K RPM 200-225 1400-1575 1600-1800 While disks may achieve these throughputs, they imply a lot of queuing delay and high response times. These ranges probably represent acceptable performance only for batch oriented applications, where throughput is the paramount performance metric. For Online Transaction Processing (OLTP) applications, these throughputs may already have unacceptably high response times. Because 15K RPM DDMs are most commonly used in OLTP environments where response time is at a premium, a simple rule of thumb is this: if the rank is doing more than 1000 Ops/sec, it is very busy, no matter what the RPM. If available, it is the average frontend response time that really matters. In addition to these enterprise class drives, near-line drives of high capacity and somewhat lower performance capabilities are now becoming options in mixtures with higher performing, enterprise class drives. These are definitely considered lower performance, capacity oriented drives, and have their own limits (see Table 3-1). Table 3-1 DDM speed and operations DDM Speed Max Ops/sec 6+P Ops/sec 7+P Ops/sec 7.2K RPM 85-110 595-770 680-880 These drive types should have limited exposure to enterprise class workloads, and the guidelines may be subject to substantial revision based on field experience. These rules of thumb or guidelines, in conjunction with knowledge of workload growth and change can be used to plan for new hardware. The discipline of capacity planning goes together with monitoring workloads and their performance. Workload characterization is just as important as performance monitoring. It is through knowledge of particular workloads and application requirements that you can focus these general guidelines into customer specific configurations. Chapter 3. Rules of thumb 35 36 Reporting with TPCTOOL 4 Chapter 4. Quick start for disk performance monitoring In this chapter we look at the disk performance metrics and available reports with TPCTOOL. © Copyright IBM Corp. 2007. All rights reserved. 37 4.1 Throughput and response time metrics Figure 4-1 Throughput metrics and response time TotalStorage Productivity V3 offers a large number of disk performance report options and each report offers useful information about the storage performance. For a quick start, let us focus on throughput and response time. To know the performance metrics available in TPC V3 Disk Manager, refer to “Commands to start” on page 22 using lstype and lsmetrics commands. You can have different levels of reports, starting from subsystem, then controller, volumes down, to device adapter and more. There are several read and write throughput metrics available for selection and inclusion in a report. Chief among these are: Total I/O Rate (overall) – includes random and sequential, read and write Read I/O Rate (overall) – includes random and sequential Write I/O Rate (overall) – includes random and sequential The corresponding response times are: Overall Response Time – average of reads and writes, including cache hits and misses Read Response Time – includes cache hits and misses Write Response Time Tip: It pays to keep historical records (and graphs) of these values over the long term. You can increase the retention period using TPC GUI. Moreover, you can periodically create text file with TPCTOOL command line and archive them using TSM or similar product. 38 Reporting with TPCTOOL 4.2 How to evaluate response time Figure 4-2 DS8000 performance - quick start It could be useful to track any growth or change in the rates and response times. It frequently happens that I/O rate grows over time and that response time increases as the I/O rates increase. As I/O rates increase, and as response times increase, you can use these trends to project when additional storage performance (as well as capacity) will be required, or alternative application designs or data layouts. It usually turns out that throughput and response time change drastically from hour to hour, day to day, and week to week. This is usually a result of different workloads between first or third shift production, or business cycles like monthend processing versus normal production.There will be periods when the values lie outside the expected range of values and the reasons will not be clear. Then the other performance metrics may be used to try to understand what is happening. The chart in Figure 4-2 is just an example to compare the response time and the throughput for two different volumes. In this case, you do not see many differences during this period between Response Time and Throughput. In other cases, you may need to split this chart in multiple charts with two or three metrics maximum. It is important to monitor the throughput and response time patterns and investigate when the numbers deviate from expected patterns. Chapter 4. Quick start for disk performance monitoring 39 The command to create that chart in is shown in Figure 4-3. C:\Program Files\IBM\TPC\cli> tpctool getrpt -user tpcadmin -pwd tpcadmin -url 9.43.85.142:9550 -columns 803,806,809,822,823,824 -level sample -subsys 2107.75BALB1+0 -ctype 12 -start 2006.10.12:03:05:35 -duration 172800 > ds8000_quickstart.txt Figure 4-3 Command to create a chart about throughput and response time The getrpt command list a performance report for a specified storage subsystem. You must have fabric operator or disk operator authority to use this command. The parameters used in this example are: -columns 803,806,809,822,823,824 specifies what columns will appear in the report. The columns are obtained from the lscounters and lsmetrics commands (see Table 4-1). Table 4-1 TPCTOOL metrics Metric Value Total Data Rate 821 Read Response Time 822 Write Response Time 823 Overall Response Time 824 Read Transfer Size 825 -subsys 2107.75BALB1+0 specifies the storage subsystem. The subsystem variable is the GUID of the storage subsystem. You can use the lsdev command to return information, including the GUID, for all storage subsystems that are discovered by TotalStorage Productivity Center. -level sample | hourly | daily specifies the level for which the performance metrics of the components should be summarized. You can specify a sample summary, an hourly summary, or a daily summary. -ctype 12 (Volume or VDisk) specifies that the output should include only components of the specified type. See the lstype command for more information about the comp_type variable. 40 Reporting with TPCTOOL 4.3 Additional metrics related to throughput Figure 4-4 Additional metrics This is a quick start document, so we cannot cover all the possible cases. But here is a short list of additional metrics (see Figure 4-4) that can be used to make sense of throughput and response time and proceed further: Total Cache Hit percentage - Percentage of cache hits for both sequential and non-sequential read and write operations, for a particular component over a time interval. Read Cache Hit Percentage - Percentage of cache hits for both sequential and non-sequential read operations, for a particular component over a time interval. NVS Full Percentage - Percentage of time that NVS space constraints caused I/O operations to be delayed, for a particular component over a time interval. (The ratio of delayed operations to total I/Os.) Important: WIth TPC V3.1.3, this metric name has changed, as well as delayed I/O rate. See Table 4-2. Table 4-2 NVS metrics Old metrics name (up to V 3.1.2) New metrics name (3.1.3 or later) NVS full percentage Write-Cache hits Percentage (overall) - 832 NVS delayed I/O rate Write-Cache delay I/O rate - 833 Chapter 4. Quick start for disk performance monitoring 41 Read Transfer Size (KB/Op) - Average number of KB per I/O for read operations, for a particular component over a time interval. Write Transfer Size (KB/Op) - Average number of KB per I/O for write operations, for a particular component over a time interval. Total Cache Hit percentage is the percentage of reads and writes that are handled by the cache without needing immediate access to the backend disk arrays. Read Cache Hit percentage focuses on Reads, since Writes are almost always recorded as cache hits. NVS refers to non-volatile storage for writes. If the NVS is full, a write may be delayed while some changed data is destaged to the disk arrays to make room for the new write data in NVS. The Read and Write Transfer Sizes are the average number of bytes transferred per I/O operation. There are many more metrics available via TPC, but these are important ones for understanding throughput and response time. Table 4-3 shows the metrics you could use for -ctype 12 at volume level, for a DSxxxx subsystem. For a complete list of all metrics available, per subsystem, refer to Chapter 7, “Metrics per subsystem” on page 85. Table 4-3 TPCTOOL metrics Metric Value ============================================== Total Data Rate 821 Read Response Time 822 Write Response Time 823 Overall Response Time 824 Read Transfer Size 825 Write Transfer Size 826 Overall Transfer Size 827 Record Mode Read I/O Rate 828 Record Mode Read Cache Hit Percentage 829 Disk to Cache Transfer Rate 830 Cache to Disk Transfer Rate 831 Write-cache Delay Percentage 832 Write-cache Delay I/O Rate 833 Read I/O Rate (overall) 803 Write I/O Rate (normal) 804 Write I/O Rate (sequential) 805 Write I/O Rate (overall) 806 Total I/O Rate (normal) 807 Total I/O Rate (sequential) 808 Total I/O Rate (overall) 809 Read Cache Hit Percentage (normal) 810 Read Cache Hits Percentage (sequential) 811 Read Cache Hits Percentage (overall) 812 Write Cache Hits Percentage (normal) 813 Write Cache Hits Percentage (sequential) 814 Write Cache Hits Percentage (overall) 815 Total Cache Hits Percentage (normal) 816 Total Cache Hits Percentage (sequential) 817 Total Cache Hits Percentage (overall) 818 Read Data Rate 819 Write Data Rate 820 Read I/O Rate (normal) 801 Read I/O Rate (sequential) 802 42 Reporting with TPCTOOL If you need to further investigate your disk performance, you may create additional reports using a command as shown in Figure 4-5. C:\Program Files\IBM\TPC\cli> tpctool getrpt -user tpcadmin -pwd tpcadmin -url 9.43.85.142:9550 -columns 818,810,832,825,826 -level sample -subsys 2107.75BALB1+0 -ctype 12 -start 2006.10.12:03:05:35 -duration 172800 > ds8000_quickstart.txt Figure 4-5 Commands to create report about cache and transfer size Chapter 4. Quick start for disk performance monitoring 43 4.4 Backend and frontend metrics Figure 4-6 Backend and frontend Throughput is measured and reported in several different ways. There is throughput for your entire Storage subsystem, or of each cluster or controller, or of each volume (or LUN). You can measure throughput at Fiber Channels interface (ports level or at the RAID ranks level, after cache hits have been filtered out. Frontend I/O metrics (see Figure 4-6) are average of all traffic between the servers and storage box and for account for relatively fast hits in the cache, as well as occasional cache misses that go all the way to the RAID ranks on the back end. Most storage boxes give metrics for both kinds of I/O operations, frontend and backend. 44 Reporting with TPCTOOL 4.5 Backend response time Figure 4-7 Ds8000 quick start backend metrics Backend Response time is the time to do staging or destaging between cache and disk arrays. The chart in Figure 4-7 can be created with the command in Figure 4-8. C:\Program Files\IBM\TPC\cli> tpctool getrpt -user tpcadmin -pwd tpcadmin -url 9.43.85.142 :9550 -columns 809,837,843 -level sample -subsys 2107.75BALB1+0 -ctype 9 -start 2006.10.16:20:20:00 -duration 86400 > ds8000_quickstart_backend.txt Figure 4-8 Command to create backend metrics reports Figure 4-9 shows the backend metrics available for a device adapter. You may get backend reports at different levels. Chapter 4. Quick start for disk performance monitoring 45 C:\Program Files\IBM\TPC\cli> tpctool lsmetrics -ctype 10 -user tpcadmin -pwd tpcadmin -url 9.43.85.142:9550 -subsys 2105.22513+0 Metric Value ============================================== Total Data Rate 821 Read Response Time 822 Write Response Time 823 Overall Response Time 824 Read Transfer Size 825 Write Transfer Size 826 Overall Transfer Size 827 Record Mode Read I/O Rate 828 Record Mode Read Cache Hit Percentage 829 Disk to Cache Transfer Rate 830 Cache to Disk Transfer Rate 831 Write-cache Delay Percentage 832 Write-cache Delay I/O Rate 833 Backend Read I/O Rate 835 Backend Write I/O Rate 836 Total Backend I/O Rate 837 Backend Read Data Rate 838 Backend Write Data Rate 839 Total Backend Data Rate 840 Backend Read Response Time 841 Backend Write Response Time 842 Overall Backend Response Time 843 Backend Read Transfer Size 847 Backend Write Transfer Size 848 Overall Backend Transfer Size 849 Read I/O Rate (overall) 803 Write I/O Rate (normal) 804 Write I/O Rate (sequential) 805 Write I/O Rate (overall) 806 Total I/O Rate (normal) 807 Total I/O Rate (sequential) 808 Total I/O Rate (overall) 809 Read Cache Hit Percentage (normal) 810 Read Cache Hits Percentage (sequential) 811 Read Cache Hits Percentage (overall) 812 Write Cache Hits Percentage (normal) 813 Write Cache Hits Percentage (sequential) 814 Write Cache Hits Percentage (overall) 815 Total Cache Hits Percentage (normal) 816 Total Cache Hits Percentage (sequential) 817 Total Cache Hits Percentage (overall) 818 Read Data Rate 819 Write Data Rate 820 Read I/O Rate (normal) 801 Read I/O Rate (sequential) 802 Disk Utilization Percentage 850 Sequential I/O Percentage 851 Figure 4-9 Backend metrics 46 Reporting with TPCTOOL Other than this, it can be useful to set some thresholds, such as: TOTAL I/O rate threshold - Sets threshold on the average number of I/O operations per second for read and write operations, for the subsystem controllers (clusters) or I/O Groups. The Total I/O Rate metric for each controller or I/O Group is checked against the threshold boundaries for each collection interval. These thresholds are disabled by default. TOTAL Backend I/O rate threshold - Sets thresholds on the average number of I/O operations per second for MDisk read and write operations, for the MDisk Groups. Or, this is the average number of I/O operations per second for read and write operations for physical volumes. The Total I/O Rate metric for each MDisk Group is checked against the threshold boundaries for each collection interval. This threshold is disabled by default. Overall Backend Response time threshold (this is the most important one) - Average number of milliseconds that it took to service each I/O operation (read and write), for a particular component over a time interval. For SVC, this is the external response time of the MDisks. Backend I/O rate is the rate of I/O between cache and disk RAID ranks in the backend of the storage. This typically include Disk Reads from the array to cache caused by a Read Miss in the cache. The Disk Write activity from cache to disk array is normally an asynchronous operation to move change data from cache to disk, freeing up space in the NVS. The Backend Response time gets averaged together with response time for cache hits, to give the Overall Response time mentioned earlier. Important: It must be always clear whether you are looking at throughput and response time at the frontend (very close to system level response time as measured from a server at operating system level) or the throughput/response time at the backend (just between cache and disk). Chapter 4. Quick start for disk performance monitoring 47 4.6 Historical performance charts Figure 4-10 DS8000 quick start - Total I/O rate This throughput chart (see Figure 4-10) summarizes the throughput by hour for a particular day. This data can be easily exported into various formats for further analysis or for archiving. You can export from the GUI a CSV file (or many other formats) or you can store directly the text file created by TPCTOOL CLI, using a command similar to the one in Figure 4-11. C:\Program Files\IBM\TPC\cli> tpctool getrpt -user tpcadmin -pwd tpcadmin -url 9.43.85.142:9550 -columns 809,837,843 -level sample -subsys 2107.75BALB1+0 -ctype 10 -start 2006.10.16:20:20:00 -duration 86400 > ds8000_quickstart_backend.txt Figure 4-11 Command to create report at Array/DS Rank level The recommendation is to build up a historical archive of performances for various subsystems, critical volumes, disk arrays and other storage subsystem resources. When there are throughput or response time anomalies, we suggest that you look at performance reports for other metrics and other resources, such as Percentage of NVS Full, 48 Reporting with TPCTOOL or the performance of individual RAID ranks, or particular volumes in critical applications as shown in Figure 4-12. Figure 4-12 DS8000 quick start Backend Response time The Response time chart (see Figure 4-12) is still related to the same DS Rank as in the example above (see Figure 4-10). The Backend Response time (and throughput) are available for all models of ESS, DS6000, DS8000 and SVC. See the command used in Figure 4-13. The chart shows the milliseconds per operations or Response time for backend reads. Read misses to cache are serviced by the backend RAID ranks and, for this rank, average 12 to 20 milliseconds response time can be considered perfectly normal for backend read response times. Actual rules of thumb for response time depend strongly on workload, time of day and other factors. There is no “cookbook value” that works for every application and storage configuration. C:\Program Files\IBM\TPC\cli> tpctool getrpt -user tpcadmin -pwd tpcadmin -url 9.43.85.142:9550 -columns 841 -level sample -subsys 2107.75BALB1+0 -ctype 10 -start 2006.10.16:20:20:00 -duration 86400 > ds8000_quickstart_backend_read.txt Figure 4-13 command used to create Backend Response time report The key is to monitor normal operations, develop an understanding of expected behavior and then track the behavior for either performance anomalies or simple growth in the workload. This historical performance information is the best source of data for capacity planning, too. Chapter 4. Quick start for disk performance monitoring 49 Retention of the performance data in the TPC database is controlled by a policy setting. After some period of time, the data is rolled up into one hour summaries. Eventually, the data is aged out of the database. So it is important to develop a set of graphs and reports to summarize and visualize the data, and to also keep the graphs and reports in some sort of historical archive. You do not need to keep every day of every month, but you do need to keep periodic snapshots of performance. In the event of performance questions, the frequency of the data samples should be increased. The number of days, weeks and/or months that performance data is retained can be modified by selecting Administrative Services → Configuration → Resource History Retention. In the screenshow below the options selected are rather large because this is a demo system. Remember that more data retained will impact the size of your TPC database. 50 Reporting with TPCTOOL 4.7 Port Data Rate and Port Response time These are useful metrics and you also can set thresholds against them, but they do not usually impact the throughput and response time from disk storage. So, usually you do not go through these in a TPC performance monitoring analysis for disk. When you create reports against Port Data Rate and Port Response Time metrics it is usually to investigate a problem in the path between the servers and storage. Chapter 4. Quick start for disk performance monitoring 51 52 Reporting with TPCTOOL 5 Chapter 5. Sample TPCTOOL reports In this chapter, we go through examples of using TPCTOOL and the GETRPT command. These reports are intended to provide you with enough information so that you can start to customize your reports according to your needs. Chapter 7, “Metrics per subsystem” on page 85 contains the detailed list of which metrics are available per each storage subsystem or fabric. For additional information and suggestions about methodology and about how to proceed, refer to Monitoring Storage Subsystems Using TotalStorage Productivity Center, SG24-7364. © Copyright IBM Corp. 2007. All rights reserved. 53 5.1 Data rate vs. response time for disk storage subsystem Figure 5-1 Data rate verses response time In this case, we collect the data (metrics values 821, 824 - see “Commands to start” on page 22) against a DS8000 volume. You could create the same charts for DS family, as well for all the other supported Storage Subsystems. Chapter 7, “Metrics per subsystem” on page 85 contains the detailed list of which metrics are available per each storage subsystem or fabric. You can check all the available metrics for each subsystem, using the command lsmetrics as described in “How to create a graph from a text file” on page 24. The chart in Figure 5-1 has been created using TPCTOOL command shown in Figure 5-2. C:\Program Files\IBM\TPC\cli>tpctool getrpt -user tpcadmin -pwd tpcadmin -ctype subsystem -url 9.43.85.142:9550 -subsys 2107.75BALB1+0 -level sample -columns 821,824 -duration 86400 -start 2006.10.12:03:00:00 > ds8000_datarate_resptime.txt Figure 5-2 Data Rate versus Response time The getrpt command lists a performance report for a specified storage subsystem. You must have fabric operator or disk operator authority to use this command. The output is redirected to a text file. The parameters used in this example are: -columns 821,824 specifies what columns will appear in the report. The columns are obtained from the lscounters and lsmetrics commands as shown in Table 5-1 on page 55. 54 Reporting with TPCTOOL Table 5-1 I/O rate and response time metrics Metric Value Total Data Rate 821 Overall response Time 824 -subsys subsystem specifies the storage subsystem. The subsystem variable is the GUID of the storage subsystem. You can use the lsdev command to return information, including the GUID, for all storage subsystems that are discovered by TotalStorage Productivity Center. -level sample | hourly | daily specifies the level for which the performance metrics of the components should be summarized. You can specify a sample summary, an hourly summary, or a daily summary. -ctype subsystem specifies that the output should include only components of the specified type. See the lstype command for more information about the comp_type variable. -start date/time specifies the date and time to start the sampling period. The date and time are formatted as: yyyy.MM dd:HH:mm:ss All time zones are relative to the Device server. See the lstime command for more information. -duration 86400 specifies the duration of the sampling period, in seconds. The duration_seconds variable is an integer. 86400 seconds means one day. Chapter 5. Sample TPCTOOL reports 55 5.2 Ports report for a disk storage subsystem Figure 5-3 DS8000 port report The chart above (Figure 5-3) shows multiple metrics I/O rate for two PORTs of a DS8000 Storage Subsystem. Chapter 7, “Metrics per subsystem” on page 85 contains the detailed list of which metrics is available per each storage subsystem or fabric. This could help to compare traffic workload and utilization on two different ports. The command used (shown in Figure 5-4) can be customized as needed. C:\PROGRA~1\IBM\TPC\cli>tpctool getrpt -user tpcadmin -pwd tpcadmin -url 9.43.85.142:9550 -columns 852,853,854 -level sample -subsys 2107.75BALB1+0 -ctype 2 -start 2006.10.13:02:10:00 -duration 86400 > ports_ds8000.txt Figure 5-4 Port report The getrpt command lists a performance report for a specified storage subsystem. You must have fabric operator or disk operator authority to use this command. The output is redirected to a text file. The parameters used in this example are: -columns 852,853,854 specifies what columns will appear in the report. The columns are obtained from the lscounters and lsmetrics commands. 56 Reporting with TPCTOOL Table 5-2 Port metrics Metric Value Port Send I/O Rate 852 Port Receive I/O Rate 853 Total Port I/O Rate 854 -subsys subsystem specifies the storage subsystem. The subsystem variable is the GUID of the storage subsystem. You can use the lsdev command to return information, including the GUID, for all storage subsystems that are discovered by TotalStorage Productivity Center. -level sample | hourly | daily specifies the level for which the performance metrics of the components should be summarized. You can specify a sample summary, an hourly summary, or a daily summary. -ctype 2 (Subsystem port) specifies that the output should include only components of the specified type. See the lstype command for more information about the comp_type variable. -start date/time specifies the date and time to start the sampling period. The date and time are formatted as: yyyy.MM dd:HH:mm:ss All time zones are relative to the Device server. See the lstime command for more information. -duration duration_seconds specifies the duration of the sampling period, in seconds. The duration_seconds variable is an integer. Chapter 5. Sample TPCTOOL reports 57 5.3 SVC performance Figure 5-5 SVC performance reports The CLI interface provides the breadth of data to perform performance analysis on storage environments using the TPCTOOL. This is based upon the capability of providing both an interactive and batch method for creating a column based report that could be imported into a spreadsheet for analysis by a client. If you want to review performance data for an SVC and the component in question was VDisk performance, then when you executed the TPCTOOL with the VDisk alias, the output would include the performance data for a specific time and for specific metrics. With SVC Version 3.1 the SVC development team exposed the VDisk response time metric. This metric along with similar current metrics in the MDisk component allow for performance analysis at the time to wait level for a client host. A large percentage of performance problems can be diagnosed using TPC . Using TPC V3.1.3 and SVC 4.1 you can get additional metrics and component types, such as SVC Node. Refer to Chapter 7, “Metrics per subsystem” on page 85 for details. Figure 5-6 shows the command to create the chart above to get multiple performance metrics against the same VDisk. This is a good a starting point for performance analysis and then you could drill down using the GUI and using the collected data. 58 Reporting with TPCTOOL C:\PROGRA~1\IBM\TPC\cli>tpctool getrpt -url acmeurl -user myid -pwd mypwd -ctype 12 -columns 803,806,809,812,815,818,819,820,821,825,826,827,830,831,832,833 -level sample -subsys asvc1 -fs Figure 5-6 Command - SVC performance for a vDisk The getrpt command lists a performance report for a specified storage subsystem. You must have fabric operator or disk operator authority to use this command. The parameters used in this example are: -columns 803,806,809,812,815,818,819,820,821,825,826,827,830,831,832,833 specifies what columns will appear in the report. The columns are obtained from the lscounters and lsmetrics commands. Table 5-3 SVC metrics and values Metric Value Read I/O Rate (overall) 803 Write I/O Rate (overall) 806 Total I/O Rate (overall) 809 Read Cache Hits Percentage (overall) 812 Write Cache Hits Percentage (overall) 815 Total Cache Hits Percentage (overall) 818 Read Data Rate 819 Write Data Rate 820 Total Data Rate 821 Read Transfer Size 825 Write Transfer Size 826 Overall Transfer Size 827 Disk to Cache Transfer rate 830 Cache to Disk Transfer Rate 831 Write-cache Delay Percentage 832 Write-cache Delay Delay I/O Rate 833 -subsys asvc1 (this is an alias configured into TPCCLI.CONF) specifies the storage subsystem. The subsystem variable is the GUID of the storage subsystem. You can use the lsdev command to return information, including the GUID, for all storage subsystems that are discovered by TotalStorage Productivity Center. -level sample | hourly | daily specifies the level for which the performance metrics of the components should be summarized. You can specify a sample summary, an hourly summary, or a daily summary. Chapter 5. Sample TPCTOOL reports 59 -ctype 12 (Volume or VDisk) specifies that the output should include only components of the specified type. See the lstype command for more information about the comp_type variable. 60 Reporting with TPCTOOL 5.4 Reports for a switch fabric Figure 5-7 Two ports and multiple metrics Chapter 5. Sample TPCTOOL reports 61 The chart in Figure 5-7 has been created using the command in Figure 5-8. As soon as you get the output text file, using Excel or any other tools, you can choose how to custom the report. You may compare multiple metrics for two or more ports as shown in Figure 5-7 (with more than two it could become confusing). Otherwise, you may focus on a few metrics (for example, Port Peak Data rate metrics) and compare them among all switch ports, as shown in (see Figure 5-9) for the first 8 ports. This can be customized at any time, working with Excel or the tool you choose to elaborate the text file generated by tpctool getrpt command. C:\PROGRA~1\IBM\TPC\cli>tpctool getrpt -user tpcadmin -pwd tpcadmin -url 9.43.85.142:9550 -columns 855,856,857,858,859,860,861,862,869,870,871,872,873,874,875,876,877,878,879,880 ,881,882,883 -level sample -fabric 100000051E34E895 -ctype 14 -start 2006.10.11:01:21:00 -duration 86400 > ports_switch.txt Figure 5-8 command to produce report against switch ports Figure 5-9 Ports peak Data Rate The getrpt command lists a performance report for a specified storage subsystem. You must have fabric operator or disk operator authority to use this command. The parameters used in this example are: -columns 855,856,857,858,859,860,861,862,869,870,871,872,873,874,875,876,877,878,879,880 62 Reporting with TPCTOOL ,881,882,883 specifies what columns will appear in the report. The columns are obtained from the lscounters and lsmetrics commands. Table 5-4 Switch port metrics Metric Value Port Send Data Rate 858 Link Failure Rate 874 CRC Error Rate 877 Port Send Packet Rate 855 Port Receive Packet Rate 856 Total Port Packet Rate 857 Port Receive Data Rate 859 Total Port Data Rate 860 Port Peak Send Data Rate 861 Port Peak Receive Data Rate 862 Port Send Packet Size 869 Port Receive Packet Size 870 Overall Port Packet Size 871 Error Frame Rate 872 Dumped Frame Rate 873 Loss of Sync Rate 875 Loss of Signal Rate 876 Short Frame Rate 878 Long Frame Rate 879 -subsys asvc1 (this is an alias configured into TPCCLI.CONF) specifies the storage subsystem. The subsystem variable is the GUID of the storage subsystem. You can use the lsdev command to return information, including the GUID, for all storage subsystems that are discovered by TotalStorage Productivity Center. -level sample | hourly | daily specifies the level for which the performance metrics of the components should be summarized. You can specify a sample summary, an hourly summary, or a daily summary. -ctype 14 (Switch port) specifies that the output should include only components of the specified type. See the lstype command for more information about the comp_type variable. -start date/time specifies the date and time to start the sampling period. The date and time are formatted as: yyyy.MM dd:HH:mm:ss All time zones are relative to the Device server. See the lstime command for more information. -duration duration_seconds specifies the duration of the sampling period, in seconds. The duration_seconds variable is an integer. Chapter 5. Sample TPCTOOL reports 63 64 Reporting with TPCTOOL 6 Chapter 6. Macro to create charts from TPCTOOL CLI text file In this chapter we show you how to import the output data, the getrpt (rptfast4500) script that has been executed for the Subsystem performance report into Excel or a similar spreadsheet, and then to create a template for later use. This imported data will now be in a more readable format for analysis and for creating reports and graphs. Chapter 7, “Metrics per subsystem” on page 85 contains the detailed list of which metrics are available per each storage subsystem or fabric. © Copyright IBM Corp. 2007. All rights reserved. 65 6.1 Importing and exporting data Data may need to be exported or imported regularly. In this case, the data has to be exported to the rptfast4500.out file and then read by the application (MS Excel). Alternatively, you can copy data on an ad hoc basis. The following example was built using Microsoft® Excel office 2000. The first task is to import the data into Excel. Follow these tasks. 1. Open a new Excel document (see Figure 6-1). Figure 6-1 A new excel document 2. From the Data drop-down menu box, select Import External Data (see Figure 6-2). 66 Reporting with TPCTOOL Figure 6-2 Select Data → Import External Data → Import Data 3. Locate the data file in the directory where you stored the TPC CLI output file (rptfast4500.out). Once selected, press Open and start the import process (see Figure 6-3). Chapter 6. Macro to create charts from TPCTOOL CLI text file 67 Figure 6-3 Select the data source to begin the import process 4. This will start the text import wizard; select Delimited (default) and Next (see Figure 6-4). Figure 6-4 Select Delimited (default) and Next 5. Select Tab (default) and enter the delimiter you selected in your script. For our example, we selected semicolon (;) as shown in Example 6-1. 68 Reporting with TPCTOOL Example 6-1 The delimiter chosen in our script -fs getrpt -user myuser -pwd mypass -url myurl -subsys fast4500 -level sample -ctype subsystem -columns 803,806,809,812,815,818,819,820,821,825,826,827 -start 2006.09.19:16:21:00 -duration 80000 -fs ; Figure 6-5 Select delimiter semicolon 6. Once the delimiter has been selected, press Next → Finish (see Figure 6-6). Figure 6-6 Select Finish to complete Wizard 7. Press OK to complete the task (see Figure 6-7). Chapter 6. Macro to create charts from TPCTOOL CLI text file 69 Figure 6-7 Press OK to complete the task Figure 6-8 Excel spreadsheet with columns unformatted The Excel document will need to be formatted to allow a template to be created. First, you need to delete row 2 and then rename the metrics to more understandable headings. These metrics heading are obtainable for using the lsmetrics command. The Excel spreadsheet result is similar to Figure 6-9. 70 Reporting with TPCTOOL Figure 6-9 Excel spreadsheet with columns renamed. You need to copy row 1 (Header Row) and paste it into a new book; the reason for this is if you attempt to delete the data below row 1 within Excel and save it as a template, then upon reusing Excel it will prompt you to refresh the data from the original source file. Therefore you need to copy the header row into a new book. Once you have copied the header row into a new book it can be saved as a template. 6.1.1 TimeStamp This extension inserts the current date/time into a message or an input field in the browser.Timestamp can refer to a time code or to a digitally signed timestamp. Timestamps are very useful for logging events. Date, time, and their variants differ more than any other data types when you compare formats between DATETIME and DATE (see Example 6-2). Example 6-2 timestamp variants 2005-05-08 10:45 Sat June 29 23:16:57 2005 2005.08.03:10:45 The international standard date notation is YYYY-MM-DD. Where YYYY is the year, MM is the month of the year between 01 (January) and 12 (December), and DD is the day of the month between 01 and 31. For example, the third day of August in the year 1980 is written in the standard notation as1980-08-03. Chapter 6. Macro to create charts from TPCTOOL CLI text file 71 The international standard notation for the time of day is hh:mm:ss. Where hh is the number of complete hours that have passed since midnight (00-24), mm is the number of complete minutes that have passed since the start of the hour (00-59), and ss is the number of complete seconds since the start of the minute (00-60). If the hour value is 24, then the minute and second values must be zero. For example, time is 23:59:59 which represents the time one second before midnight. All of the time comparison procedures require the time objects to be of the same type. It is an error to use these procedures on time objects of different types. For the timestamp measurements we would need to convert the timestamp to meet international standard regardless of format used per country or region. We have developed a Visual Basic® Script (vbs) to collect the format used by your workstation and convert the timestamp to an international format (see Example 6-3). Example 6-3 Visual Basic Script to convert timestamp to international format ' Get Date and Time Separator String DS = Application.International(xlDateSeparator) TS = Application.International(xlTimeSeparator) If Application.International(xl24HourClock) Then AMPM = "" Else AMPM = " AM/PM" End If ' This loop runs until there is nothing in the next column Dim TimeStamp As String Do If Application.International(xlDateOrder) = 0 Then ActiveCell.NumberFormat = "mm" + DS + "dd" + DS + "yyyy hh" + TS + "mm" + TS + "ss" + AMPM ElseIf Application.International(xlDateOrder) = 1 Then ActiveCell.NumberFormat = "dd" + DS + "mm" + DS + "yyyy hh" + TS + "mm" + TS + "ss" + AMPM ElseIf Application.International(xlDateOrder) = 2 Then ActiveCell.NumberFormat = "yyyy" + DS + "mm" + DS + "dd hh" + TS + "mm" + TS + "ss" + AMPM End If TimeStamp = Replace(ActiveCell.Value, ":", " ", 1, 1, vbTextCompare) TimeStamp = Replace(TimeStamp, ".", DS, 1, 2, vbTextCompare) TimeStamp = Replace(TimeStamp, ":", TS, 1, 2, vbTextCompare) ActiveCell.Value = TimeStamp ActiveCell.Offset(1, 0).Select Loop Until IsEmpty(ActiveCell) ENDE: 72 Reporting with TPCTOOL Application.ScreenUpdating = True Application.EnableEvents = True 6.1.2 Create macros You will need to copy this VB code into your Excel worksheet and save it a macro. A macro automates a complex task. Excel macros can perform complicated series of actions or simply record commonly used commands. Using the code above we can fully automate the Timestamp conversion. These are the steps to create a macro. 1. Open the Tools menu and select Macro, and then select Macros (see Figure 6-10). Figure 6-10 Select Macro 2. Fill in the macro name, and press Create. The macro should be given a descriptive name; in this case we are using Timestamp. The macro can be available from only one worksheet or from any worksheet (see Figure 6-11). Chapter 6. Macro to create charts from TPCTOOL CLI text file 73 Figure 6-11 Macro Name 3. Copy the code in Example 6-3 on page 72 into the “Microsoft Visual Basic” editor between the “Sub” and the “End Sub” lines (see Figure 6-12). Figure 6-12 Copy Macro code in VB Editor 4. Once the code has been copied into the editor, close the editor window, which brings you back into the Excel worksheet and the macro is saved automatically. 5. Now you have successfully created a macro to convert the timestamp into an international standard. 74 Reporting with TPCTOOL How to run the macro Once the data has been imported into your worksheet and you want to convert the timestamp into an international standard, and to use it as a true value, you can run the macro. The macro is very simple to run. 1. Select Tools → Macro → Macros (Alt- F8) as shown in Figure 6-13. Figure 6-13 Browse to run the Macro 2. Select the macro you want to run. In this case, we selected “Timestamp” and then pressed Run (see Figure 6-14). Chapter 6. Macro to create charts from TPCTOOL CLI text file 75 Figure 6-14 Select the macro to run 3. This will format the timestamp column (see Figure 6-15). Figure 6-15 Timestamp column formatted You may save this as a complete template which includes the Performance report headings and the Timestamp convertor. 76 Reporting with TPCTOOL 6.1.3 Creating a template Using the Excel spreadsheet above you need to create a template for later use either with or without the macro; this is dependent on the type of graphs you want to create (see Figure 6-16). Provide a descriptive name for the template. Figure 6-16 Save the book as a template for reuse Important: Select Template (*.xlt) from the Save as type drop-down box. Once you have saved this template you may import performance extracted data into this spreadsheet. Restriction: Only the same data criteria can be imported into the same Template, for example, DS4000 Subsystem metrics. The ITSO has created a template for you to import your extracted data into. The template headings may be modified to compensate for other metrics and reports, however the macro and the layout has been created for you. This template can downloaded from the ITSO Web site. Refer to the Redbooks publication TotalStorage Productivity Center Advanced Topics, SG24-7348, download materials. Importing into a saved template Use the same methodology described above import the performance extracted data into the saved template. 1. Open the saved template. 2. Select the source file. Chapter 6. Macro to create charts from TPCTOOL CLI text file 77 3. Start the import wizard. 4. For Step 1 of 3 (see Figure 6-17), in the field Start import at row:, choose 3. This will remove the header from the source file and the additional separation line. Figure 6-17 Excel - change the start import row number 5. Choose the delimiter used when extracting the data in Step 2 of 3. 6. Select Finish. This will return an output similar to Figure 6-18. Figure 6-18 Final output of imported data to a template 78 Reporting with TPCTOOL Now the data extracted from TPC will be available to be used for analysis. Furthermore, this data could be used to create graphs. Depending on the request and the analysis required, you may convert the timestamp into an international standard or use it as a string. 6.1.4 Creating graphs Using the data that has been extracted from TPC and imported into the Excel template you will now be able to create readable graphs. Graphs are used to determine relationships by plotting large numbers of data points and observing the grouping or clustering of the data points. Templates help you create a macro to copy data from the Excel sheet to any application. Excel macros can be recorded so these templates for creating graphs can be reused in the future for reproducing graphs of the same type; in our example, Subsystem reports for DS4500. After creating your template and importing the data extracted from TPC via the TPCTOOL, you need to determine what type of report you want to generate. In this example, we show you how to create a performance report using the timestamp, Read I/O, Write I/O and Total I/O. Note: For this example we used the macro to convert the timestamp into an international standard and use the timestamp as a data value and not as a string. 1. Select the columns you want to plot onto the graph, that is, timestamp, Read I/O, Write I/O and Total I/O (see Figure 6-19). Note we are only selecting a few lines of entries of data. Figure 6-19 Select Columns to create a graph 2. You need to utilize the chart wizard to plot the values onto the graph. Select Insert → Chart Wizard (see Figure 6-20). Chapter 6. Macro to create charts from TPCTOOL CLI text file 79 Figure 6-20 Begin Chart Wizard 3. Select the chart type you want to use to create the graph. Graphs (line graphs and scatter plots) Line graphs provide an excellent way to map independent and dependent variables that are both quantitative. When both variables are quantitative, the line segment that connects two points on the graph expresses a slope, which can be interpreted visually relative to the slope of other lines or expressed as a precise mathematical formula. Scatter plots are similar to line graphs in that they start with mapping quantitative data points. The difference is that with a scatter plot, the decision is made that the individual points should not be connected directly together with a line, but instead express a trend. This trend can be seen directly through the distribution of points or with the addition of a regression line. A statistical tool is used to mathematically express a trend in the data. Tip: Using the XY scatter plot graph will provide you with the most realistic graph as both axis X and Y would be true value plot points on the graph, instead of points. However the graphs maybe be produced in any format for analysis. The graph types produced are dependent on the analysis you want to achieve. We use the XY scatter graph to analyze and produce the graph (Step 1 of 4). See Figure 6-21. 80 Reporting with TPCTOOL Figure 6-21 Selecting XY Scatter Graph 4. Select Next once the graph type is chosen. The next window will confirm the data range (Step 2 of 4). Click Next (Step 3 of 4) as shown in Figure 6-22. Figure 6-22 Step 3 of 4- Chart Options 5. You need to enter the chart title, X value axis title and the Y value axis title. Descriptive titles are used for each graph (see Figure 6-23). Chapter 6. Macro to create charts from TPCTOOL CLI text file 81 Figure 6-23 Input chart titles 6. The next panel (Step 4 of 4) allows you to create the graph in the same sheet or in a new sheet. This is entirely up to you. If you are producing multiple graphs from the same template, then we recommend that you place each graph in a new sheet. In this example we use the same sheet. Now the graph will be generated (see Figure 6-24). Figure 6-24 Performance Graph for Subsystem 82 Reporting with TPCTOOL Looking at the performance graph in Figure 6-24 on page 82, we notice a consistent I/O rate throughout the time frame. For an I/O graph like this, we look for spikes and drill deeper into the subsystem to determine what causes these spikes. As described above, it is easy to use Excel to produce performance graphs. Performance graphs can similarly produce graphs for other subsystems, switch, and SAN Volume Controller (SVC). For more information about producing custom performance reports, refer to the paper titled: A Quickstart to TotalStorage Productivity Center Performance Reporting One or more of the following URL links will show you the document: – IBM http://w3.ibm.com/support/techdocs/atsmastr.nsf/WebIndex/WP100794 – Business Partners http://partners.boulder.ibm.com/src/atsmastr.nsf/WebIndex/WP100794 Link to IBM TotalStorage Productivity Center V3 - Performance Management Best Practices: http://www-1.ibm.com/support/docview.wss?uid=ssg1S7001493&rs=1133 Chapter 6. Macro to create charts from TPCTOOL CLI text file 83 84 Reporting with TPCTOOL 7 Chapter 7. Metrics per subsystem In this chapter we provide the available metrics per subsystem that can be used to generate reports using TPCTOOL. © Copyright IBM Corp. 2007. All rights reserved. 85 7.1 Metrics for DS4000 storage subsystem Table 7-1 lists the DS4000 component type available for reports. In TPCTOOL reports you can specify either -ctype subsystem or -ctype 1 as a metric and get the same result. Table 7-1 DS4000 storage subsystem component Component Type Available subsystem 1 Subsystem subsys_port 2 HBA port vol 12 Volume Table 7-2 lists the DS4000 metrics that can be reported on by subsystem. Table 7-2 Metrics for DS4000 ctype subsystem DS4000 ctype: subsystem 1 (Subsystem) Metric Value Write Data Rate820 820 Total Data Rate 821 Read Transfer Size 825 Write Transfer Size 826 Overall Transfer Size 827 Read I/O Rate (overall) 803 Write I/O Rate (overall) 806 Total I/O Rate (overall) 809 Write Cache Hits Percentage (overall) 815 Total Cache Hits Percentage (overall) 818 Read Data Rate 819 Table 7-3 lists the DS4000 metrics that can be reported on by HBA port. Table 7-3 Metrics for DS4000 ctype sub_port2 DS4000 - ctype: subsys_port 2 86 (HBA port) Metric Value Total Port I/O Rate 854 Total Port Data Rate 860 Total Port Transfer Size 868 Reporting with TPCTOOL Table 7-4 lists the DS4000 metrics that can be reported on By Volume. Table 7-4 Metrics for DS4000 ctype VolumeComponent DS400 - ctype: vol 12 (VolumeComponent) Metric Value Write Data Rate 820 Total Data Rate 821 Read Transfer Size 825 Write Transfer Size 826 Overall Transfer Size 827 Read I/O Rate (overall) 803 Write I/O Rate (overall) 806 Total I/O Rate (overall) 809 Read Cache Hits Percentage (overall) 812 Write Cache Hits Percentage (overall) 815 Total Cache Hits Percentage (overall) 818 Read Data Rate 819 7.2 Metrics for ESS storage subsystem Table 7-5 lists the ESS component types available for reports. Table 7-5 ESS Storage Subsystems components Component type available subsystem 1 Subsystem subsys_port 2 HBA port controller 3 Controller da 8 Device Adapter array 10 Array vol 12 VolumeComponent Table 7-6 lists the ESS metrics by subsystem. Table 7-6 Metrics for ESS by subsystem ESS - ctype: subsystem 1 (Subsystem) Metric Value Total Data Rate 821 Read Response Time 822 Write Response Time 823 Chapter 7. Metrics per subsystem 87 ESS - ctype: subsystem 1 (Subsystem) 88 Overall Response Time 824 Read Transfer Size 825 Write Transfer Size 826 Overall Transfer Size 827 Record Mode Read I/O Rate 828 Record Mode Read Cache Hit Percentage 829 Disk to Cache Transfer Rate 830 Cache to Disk Transfer Rate 831 Write-cache Delay Percentage 832 Write-cache Delay I/O Rate 833 Cache Holding Time 834 Backend Read I/O Rate 835 Backend Write I/O Rate 836 Total Backend I/O Rate 837 Backend Read Data Rate 838 Backend Write Data Rate 839 Total Backend Data Rate 840 Backend Read Response Time 841 Backend Write Response Time 842 Overall Backend Response Time 843 Port Receive Data Rate 859 Backend Read Transfer Size 847 Backend Write Transfer Size 848 Overall Backend Transfer Size 849 Read I/O Rate (overall) 803 Write I/O Rate (normal) 804 Write I/O Rate (sequential) 805 Write I/O Rate (overall) 806 Total I/O Rate (normal) 807 Total I/O Rate (sequential) 808 Total I/O Rate (overall) 809 Read Cache Hits Percentage (normal) 810 Read Cache Hits Percentage (sequential) 811 Read Cache Hits Percentage (overall) 812 Reporting with TPCTOOL ESS - ctype: subsystem 1 (Subsystem) Write Cache Hits Percentage (normal) 813 Write Cache Hits Percentage (sequential) 814 Write Cache Hits Percentage (overall) 815 Total Cache Hits Percentage (normal) 816 Total Cache Hits Percentage (sequential) 817 Total Cache Hits Percentage (overall) 818 Read Data Rate 819 Write Data Rate 820 Read I/O Rate (normal) 801 Read I/O Rate (sequential) 802 Port Send I/O Rate 852 Port Receive I/O Rate 853 Total Port I/O Rate 854 Port Send Data Rate 858 Total Port Data Rate 860 Port Send Response Time 863 Port Receive Response Time 864 Total Port Response Time 865 Port Send Transfer Size 866 Port Receive Transfer Size 867 Total Port Transfer Size 868 Table 7-7 lists the ESS metrics by HBA port. Table 7-7 ESS metrics by HBA port ESS - ctype:subsys_port 2 (HBA port) Metric Value Port Receive Data Rate 859 Port Send I/O Rate 852 Port Receive I/O Rate 853 Total Port I/O Rate 854 Port Send Data Rate 858 Total Port Data Rate 860 Port Send Response Time 863 Port Receive Response Time 864 Chapter 7. Metrics per subsystem 89 ESS - ctype:subsys_port 2 (HBA port) Total Port Response Time 865 Port Send Transfer Size 866 Port Receive Transfer Size 867 Total Port Transfer Size 868 Table 7-8 lists the ESS metrics by controller. Table 7-8 ESS metrics by controller ESS - ctype:controller 3 90 (Controller) Metric Value Total Data Rate 821 Read Response Time 822 Write Response Time 823 Overall Response Time 824 Read Transfer Size 825 Write Transfer Size 826 Overall Transfer Size 827 Record Mode Read I/O Rate 828 Record Mode Read Cache Hit Percentage 829 Disk to Cache Transfer Rate 830 Cache to Disk Transfer Rate 831 Write-cache Delay Percentage 832 Write-cache Delay I/O Rate 833 Cache Holding Time 834 Backend Read I/O Rate 835 Backend Write I/O Rate 836 Total Backend I/O Rate 837 Backend Read Data Rate 838 Backend Write Data Rate 839 Total Backend Data Rate 840 Backend Read Response Time 841 Backend Write Response Time 842 Overall Backend Response Time 843 Backend Read Transfer Size 847 Backend Write Transfer Size 848 Reporting with TPCTOOL ESS - ctype:controller 3 (Controller) Overall Backend Transfer Size 849 Read I/O Rate (overall) 803 Write I/O Rate (normal) 804 Write I/O Rate (sequential) 805 Write I/O Rate (overall) 806 Total I/O Rate (normal) 807 Total I/O Rate (sequential) 808 Total I/O Rate (overall) 809 Read Cache Hit Percentage (normal) 810 Read Cache Hits Percentage (sequential) 811 Read Cache Hits Percentage (overall) 812 Write Cache Hits Percentage (normal) 813 Write Cache Hits Percentage (sequential) 814 Write Cache Hits Percentage (overall) 815 Total Cache Hits Percentage (normal) 816 Total Cache Hits Percentage (sequential) 817 Total Cache Hits Percentage (overall) 818 Read Data Rate 819 Write Data Rate 820 Read I/O Rate (normal) 801 Read I/O Rate (sequential) 802 Table 7-9 lists the ESS metrics by device adapter. Table 7-9 ESS metrics by device adapter ESS - ctype: da 8 (Device Adapter) Metric Value Total Data Rate 821 Read Response Time 822 Write Response Time 823 Overall Response Time 824 Read Transfer Size 825 Write Transfer Size 826 Overall Transfer Size 827 Record Mode Read I/O Rate 828 Chapter 7. Metrics per subsystem 91 ESS - ctype: da 92 8 (Device Adapter) Record Mode Read Cache Hit Percentage 829 Disk to Cache Transfer Rate 830 Cache to Disk Transfer Rate 831 Write-cache Delay Percentage 832 Write-cache Delay I/O Rate 833 Backend Read I/O Rate 835 Backend Write I/O Rate 836 Total Backend I/O Rate 837 Backend Read Data Rate 838 Backend Write Data Rate 839 Total Backend Data Rate 840 Backend Read Response Time 841 Backend Write Response Time 842 Overall Backend Response Time 843 Backend Read Transfer Size 847 Backend Write Transfer Size 848 Overall Backend Transfer Size 849 Read I/O Rate (overall) 803 Write I/O Rate (normal) 804 Write I/O Rate (sequential) 805 Write I/O Rate (overall) 806 Total I/O Rate (normal) 807 Total I/O Rate (sequential) 808 Total I/O Rate (overall) 809 Read Cache Hit Percentage (normal) 810 Read Cache Hits Percentage (sequential) 811 Read Cache Hits Percentage (overall) 812 Write Cache Hits Percentage (normal) 813 Write Cache Hits Percentage (sequential) 814 Write Cache Hits Percentage (overall) 815 Total Cache Hits Percentage (normal) 816 Total Cache Hits Percentage (sequential) 817 Total Cache Hits Percentage (overall) 818 Read Data Rate 819 Reporting with TPCTOOL ESS - ctype: da 8 (Device Adapter) Write Data Rate 820 Read I/O Rate (normal) 801 Read I/O Rate (sequential) 802 Table 7-10 lists the ESS metrics by array. Table 7-10 ESS metrics by array ESS - ctype: array 10 ( Array ) Metric Value Total Data Rate 821 Read Response Time 822 Write Response Time 823 Overall Response Time 824 Read Transfer Size 825 Write Transfer Size 826 Overall Transfer Size 827 Record Mode Read I/O Rate 828 Record Mode Read Cache Hit Percentage 829 Disk to Cache Transfer Rate 830 Cache to Disk Transfer Rate 831 Write-cache Delay Percentage 832 Write-cache Delay I/O Rate 833 Backend Read I/O Rate 835 Backend Write I/O Rate 836 Total Backend I/O Rate 837 Backend Read Data Rate 838 Backend Write Data Rate 839 Total Backend Data Rate 840 Backend Read Response Time 841 Backend Write Response Time 842 Overall Backend Response Time 843 Backend Read Transfer Size 847 Backend Write Transfer Size 848 Overall Backend Transfer Size 849 Read I/O Rate (overall) 803 Chapter 7. Metrics per subsystem 93 ESS - ctype: array 10 ( Array ) Write I/O Rate (normal) 804 Write I/O Rate (sequential) 805 Write I/O Rate (overall) 806 Total I/O Rate (normal) 807 Total I/O Rate (sequential) 808 Total I/O Rate (overall) 809 Read Cache Hit Percentage (normal) 810 Read Cache Hits Percentage (sequential) 811 Read Cache Hits Percentage (overall) 812 Write Cache Hits Percentage (normal) 813 Write Cache Hits Percentage (sequential) 814 Write Cache Hits Percentage (overall) 815 Total Cache Hits Percentage (normal) 816 Total Cache Hits Percentage (sequential) 817 Total Cache Hits Percentage (overall) 818 Read Data Rate 819 Write Data Rate 820 Read I/O Rate (normal) 801 Read I/O Rate (sequential) 802 Disk Utilization Percentage 850 Sequential I/O Percentage 851 Table 7-11 lists the ESS metrics By Volume component. Table 7-11 ESS metrics By Volume component ESS - ctype:vol 94 12 (VolumeComponent) Metric Value Total Data Rate 821 Read Response Time 822 Write Response Time 823 Overall Response Time 824 Read Transfer Size 825 Write Transfer Size 826 Overall Transfer Size 827 Record Mode Read I/O Rate 828 Reporting with TPCTOOL ESS - ctype:vol 12 (VolumeComponent) Record Mode Read Cache Hit Percentage 829 Disk to Cache Transfer Rate 830 Cache to Disk Transfer Rate 831 Write-cache Delay Percentage 832 Write-cache Delay I/O Rate 833 Read I/O Rate (overall) 803 Write I/O Rate (normal) 804 Write I/O Rate (sequential) 805 Write I/O Rate (overall) 806 Total I/O Rate (normal) 807 Total I/O Rate (sequential) 808 Total I/O Rate (overall) 809 Read Cache Hit Percentage (normal) 810 Read Cache Hits Percentage (sequential) 811 Read Cache Hits Percentage (overall) 812 Write Cache Hits Percentage (normal) 813 Write Cache Hits Percentage (sequential) 814 Write Cache Hits Percentage (overall) 815 Total Cache Hits Percentage (normal) 816 Total Cache Hits Percentage (sequential) 817 Total Cache Hits Percentage (overall) 818 Read Data Rate 819 Write Data Rate 820 Read I/O Rate (normal) 801 Read I/O Rate (sequential) 802 7.3 Metrics for DS8000/DS6000 storage subsystems Table 7-12 lists the DS8000/DS6000 component types available for reports. Table 7-12 DS8000/DS6000 Storage Subsystem components Component types available subsystem 1 Subsystem subsys_port 2 HBA port controller 3 Controller Chapter 7. Metrics per subsystem 95 Component types available stor_pool 4 Storage Pool ds_rio 6 RIO Loop da 8 Device Adapter ds_rank 9 Rank array 10 Array vol 12 VolumeComponent Table 7-13 lists the DS8000/DS6000 metrics by subsystem. Table 7-13 DS8000/DS6000 metrics by subsystem DS8K/DS6K - ctype:subsystem 1 96 (Subsystem) Metric Value Total Data Rate 821 Read Response Time 822 Write Response Time 823 Overall Response Time 824 Read Transfer Size 825 Write Transfer Size 826 Overall Transfer Size 827 Record Mode Read I/O Rate 828 Record Mode Read Cache Hit Percentage 829 Disk to Cache Transfer Rate 830 Cache to Disk Transfer Rate 831 Write-cache Delay Percentage 832 Write-cache Delay I/O Rate 833 Cache Holding Time 834 Backend Read I/O Rate 835 Backend Write I/O Rate 836 Total Backend I/O Rate 837 Backend Read Data Rate 838 Backend Write Data Rate 839 Total Backend Data Rate 840 Backend Read Response Time 841 Backend Write Response Time 842 Overall Backend Response Time 843 Reporting with TPCTOOL DS8K/DS6K - ctype:subsystem 1 (Subsystem) Port Receive Data Rate 859 Backend Read Transfer Size 847 Backend Write Transfer Size 848 Overall Backend Transfer Size 849 Read I/O Rate (overall) 803 Write I/O Rate (normal) 804 Write I/O Rate (sequential) 805 Write I/O Rate (overall) 806 Total I/O Rate (normal) 807 Total I/O Rate (sequential) 808 Total I/O Rate (overall) 809 Read Cache Hit Percentage (normal) 810 Read Cache Hits Percentage (sequential) 811 Read Cache Hits Percentage (overall) 812 Write Cache Hits Percentage (normal) 813 Write Cache Hits Percentage (sequential) 814 Write Cache Hits Percentage (overall) 815 Total Cache Hits Percentage (normal) 816 Total Cache Hits Percentage (sequential) 817 Total Cache Hits Percentage (overall) 818 Read Data Rate 819 Write Data Rate 820 Read I/O Rate (normal) 801 Read I/O Rate (sequential) 802 Port Send I/O Rate 852 Port Receive I/O Rate 853 Total Port I/O Rate 854 Port Send Data Rate 858 Total Port Data Rate 860 Port Send Response Time 863 Port Receive Response Time 864 Total Port Response Time 865 Chapter 7. Metrics per subsystem 97 DS8K/DS6K - ctype:subsystem 1 (Subsystem) Port Send Transfer Size 866 Port Receive Transfer Size 867 Total Port Transfer Size 868 Table 7-14 lists the DS8000/DS6000 metrics by HBA port. Table 7-14 DS8000/DS6000 metrics by HBA port DS8k/DS6k - ctype:subsys_port 2 (HBA port) Metric Value Port Receive Data Rate 859 Port Send I/O Rate 852 Port Receive I/O Rate 853 Total Port I/O Rate 854 Port Send Data Rate 858 Total Port Data Rate 860 Port Send Response Time 863 Port Receive Response Time 864 Total Port Response Time 865 Port Send Transfer Size 866 Port Receive Transfer Size 867 Total Port Transfer Size 868 Table 7-15 lists the DS8000/DS6000 metrics by controller. Table 7-15 DS8000/DS6000 metrics by controller DS8K/DS6K - ctype:controller 3 98 (Controller) Metric Value Port Receive Data Rate 859 Port Send I/O Rate 852 Port Receive I/O Rate 853 Total Port I/O Rate 854 Port Send Data Rate 858 Total Port Data Rate 860 Port Send Response Time 863 Port Receive Response Time 864 Total Port Response Time 865 Reporting with TPCTOOL DS8K/DS6K - ctype:controller 3 (Controller) Port Send Transfer Size 866 Port Receive Transfer Size 867 Total Port Transfer Size 868 Table 7-16 lists the DS8000/DS6000 metrics by device adapter. Table 7-16 DS8000/DS6000 metrics by device adapter DS8K/DS6K - ctype:da 8 (Device Adapter) Metric Value Total Data Rate 821 Read Response Time 822 Write Response Time 823 Overall Response Time 824 Read Transfer Size 825 Write Transfer Size 826 Overall Transfer Size 827 Record Mode Read I/O Rate 828 Record Mode Read Cache Hit Percentage 829 Disk to Cache Transfer Rate 830 Cache to Disk Transfer Rate 831 Write-cache Delay Percentage 832 Write-cache Delay I/O Rate 833 Backend Read I/O Rate 835 Backend Write I/O Rate 836 Total Backend I/O Rate 837 Backend Read Data Rate 838 Backend Write Data Rate 839 Total Backend Data Rate 840 Backend Read Response Time 841 Backend Write Response Time 842 Overall Backend Response Time 843 Backend Read Transfer Size 847 Backend Write Transfer Size 848 Overall Backend Transfer Size 849 Read I/O Rate (overall) 803 Chapter 7. Metrics per subsystem 99 DS8K/DS6K - ctype:da 8 (Device Adapter) Write I/O Rate (normal) 804 Write I/O Rate (sequential) 805 Write I/O Rate (overall) 806 Total I/O Rate (normal) 807 Total I/O Rate (sequential) 808 Total I/O Rate (overall) 809 Read Cache Hit Percentage (normal) 810 Read Cache Hits Percentage (sequential) 811 Read Cache Hits Percentage (overall) 812 Write Cache Hits Percentage (normal) 813 Write Cache Hits Percentage (sequential) 814 Write Cache Hits Percentage (overall) 815 Total Cache Hits Percentage (normal) 816 Total Cache Hits Percentage (sequential) 817 Total Cache Hits Percentage (overall) 818 Read Data Rate 819 Write Data Rate 820 Read I/O Rate (normal) 801 Read I/O Rate (sequential) 802 Table 7-17 lists the DS8000/DS6000 metrics by rank. Table 7-17 DS8000/DS6000 metrics by rank DS8K/DS6K - ctype:ds_rank 100 9 (Rank) Metric Value Backend Read I/O Rate 835 Backend Write I/O Rate 836 Total Backend I/O Rate 837 Backend Read Data Rate 838 Backend Write Data Rate 839 Total Backend Data Rate 840 Backend Read Response Time 841 Backend Write Response Time 842 Overall Backend Response Time 843 Backend Read Transfer Size 847 Reporting with TPCTOOL DS8K/DS6K - ctype:ds_rank 9 (Rank) Backend Write Transfer Size 848 Overall Backend Transfer Size 849 Table 7-18 lists the DS8000/DS6000 metrics by array. Table 7-18 DS8000/DS6000 metrics by array DS8K/DS6K- ctype:array 10 (Array) Metric Value Total Data Rate 821 Read Response Time 822 Write Response Time 823 Overall Response Time 824 Read Transfer Size 825 Write Transfer Size 826 Overall Transfer Size 827 Record Mode Read I/O Rate 828 Record Mode Read Cache Hit Percentage 829 Disk to Cache Transfer Rate 830 Cache to Disk Transfer Rate 831 Write-cache Delay Percentage 832 Write-cache Delay I/O Rate 833 Backend Read I/O Rate 835 Backend Write I/O Rate 836 Total Backend I/O Rate 837 Backend Read Data Rate 838 Backend Write Data Rate 839 Total Backend Data Rate 840 Backend Read Response Time 841 Backend Write Response Time 842 Overall Backend Response Time 843 Backend Read Transfer Size 847 Backend Write Transfer Size 848 Overall Backend Transfer Size 849 Read I/O Rate (overall) 803 Write I/O Rate (normal) 804 Chapter 7. Metrics per subsystem 101 DS8K/DS6K- ctype:array 10 (Array) Write I/O Rate (sequential) 805 Write I/O Rate (overall) 806 Total I/O Rate (normal) 807 Total I/O Rate (sequential) 808 Total I/O Rate (overall) 809 Read Cache Hit Percentage (normal) 810 Read Cache Hits Percentage (sequential) 811 Read Cache Hits Percentage (overall) 812 Write Cache Hits Percentage (normal) 813 Write Cache Hits Percentage (sequential) 814 Write Cache Hits Percentage (overall) 815 Total Cache Hits Percentage (normal) 816 Total Cache Hits Percentage (sequential) 817 Total Cache Hits Percentage (overall) 818 Read Data Rate 819 Write Data Rate 820 Read I/O Rate (normal) 801 Read I/O Rate (sequential) 802 Disk Utilization Percentage 850 Sequential I/O Percentage 851 Table 7-19 lists the DS8000/DS6000 metrics By Volume component. Table 7-19 DS8000/DS6000 metrics By Volume component DS8K/DS6K - ctype:vol 102 12 (VolumeComponent) Metric Value Total Data Rate 821 Read Response Time 822 Write Response Time 823 Overall Response Time 824 Read Transfer Size 825 Write Transfer Size 826 Overall Transfer Size 827 Record Mode Read I/O Rate 828 Record Mode Read Cache Hit Percentage 829 Reporting with TPCTOOL DS8K/DS6K - ctype:vol 12 (VolumeComponent) Disk to Cache Transfer Rate 830 Cache to Disk Transfer Rate 831 Write-cache Delay Percentage 832 Write-cache Delay I/O Rate 833 Read I/O Rate (overall) 803 Write I/O Rate (normal) 804 Write I/O Rate (sequential) 805 Write I/O Rate (overall) 806 Total I/O Rate (normal) 807 Total I/O Rate (sequential) 808 Total I/O Rate (overall) 809 Read Cache Hit Percentage (normal) 810 Read Cache Hits Percentage (sequential) 811 Read Cache Hits Percentage (overall) 812 Write Cache Hits Percentage (normal) 813 Write Cache Hits Percentage (sequential) 814 Write Cache Hits Percentage (overall) 815 Total Cache Hits Percentage (normal) 816 Total Cache Hits Percentage (sequential) 817 Total Cache Hits Percentage (overall) 818 Read Data Rate 819 Write Data Rate 820 Read I/O Rate (normal) 801 Read I/O Rate (sequential) 802 7.4 Metrics for SVC storage subsystems Table 7-20 lists the SVC component types available for reports. Table 7-20 SVC component types available for reports Component type available subsystem 1 Subsystem subsys_port 2 HBA port svc_iogrp 5 SVC I/O Group svc_mdgrp 7 SVC Managed Disk Group Chapter 7. Metrics per subsystem 103 Component type available svc_mdisk 11 SVC Managed Disk vol 12 VolumeComponent svc_node 15 SVC Node Table 7-21lists the SVC metrics by subsystem. Table 7-21 SVC metrics by subsystem SVC - ctype:subsystem 1 104 (Subsystem) Metric Value Total Data Rate 821 Read Response Time 822 Port Receive Data Rate 859 Write Response Time 823 Overall Response Time 824 Read Transfer Size 825 Write Transfer Size 826 Overall Transfer Size 827 Disk to Cache Transfer Rate 830 Cache to Disk Transfer Rate 831 Write-cache Delay Percentage 832 Write-cache Delay I/O Rate 833 Backend Read I/O Rate 835 Backend Write I/O Rate 836 Total Backend I/O Rate 837 Backend Read Data Rate 838 Backend Write Data Rate 839 Total Backend Data Rate 840 Backend Read Response Time 841 Backend Write Response Time 842 Overall Backend Response Time 843 Read Queue Time 844 Write Queue Time 845 Overall Queue Time 846 Backend Read Transfer Size 847 Backend Write Transfer Size 848 Reporting with TPCTOOL SVC - ctype:subsystem 1 (Subsystem) Overall Backend Transfer Size 849 Port Send Data Rate 858 Read I/O Rate (overall) 803 Write I/O Rate (overall) 806 Total I/O Rate (overall) 809 Read Cache Hits Percentage (overall) 812 Write Cache Hits Percentage (overall) 815 Total Cache Hits Percentage (overall) 818 Read Data Rate 819 Write Data Rate 820 Port Send I/O Rate 852 Port Receive I/O Rate 853 Total Port I/O Rate 854 Total Port Data Rate 860 Readahead Percentage of Cache Hits 890 Dirty Write Percentage of Cache Hits 891 Write Cache Overflow Percentage 894 Write Cache Overflow I/O Rate 895 Write Cache Flush-through Percentage 896 Write Cache Flush-through I/O Rate 897 Write Cache Write-through Percentage 898 Write Cache Write-through I/O Rate 899 CPU Utilization Percentage 900 Port to Host Send I/O Rate 901 Port to Host Receive I/O Rate 902 Total Port to Host I/O Rate 903 Port to Disk Send I/O Rate 904 Port to Disk Receive I/O Rate 905 Total Port to Disk I/O Rate 906 Port to Local Node Send I/O Rate 907 Port to Local Node Receive I/O Rate 908 Total Port to Local Node I/O Rate 909 Port to Remote Node Send I/O Rate 910 Port to Remote Node Receive I/O Rate 911 Chapter 7. Metrics per subsystem 105 SVC - ctype:subsystem 1 (Subsystem) Total Port to Remote Node I/O Rate 912 Port to Host Send Data Rate 913 Port to Host Receive Data Rate 914 Total Port to Host Data Rate 915 Port to Disk Send Data Rate 916 Port to Disk Receive Data Rate 917 Total Port to Disk Data Rate 918 Port to Local Node Send Data Rate 919 Port to Local Node Receive Data Rate 920 Total Port to Local Node Data Rate 921 Port to Remote Node Send Data Rate 922 Port to Remote Node Receive Data Rate 923 Total Port to Remote Node Data Rate 924 Port to Local Node Send Response Time 925 Port to Local Node Receive Response Time 926 Overall Port to Local Node Response Time 927 Port to Local Node Send Queue Time 928 Port to Local Node Receive Queue Time 929 Overall Port to Local Node Queue Time 930 Port to Remote Node Send Response Time 931 Port to Remote Node Receive Response Time 932 Overall Port to Remote Node Response Time 933 Port to Remote Node Send Queue Time 934 Port to Remote Node Receive Queue Time 935 Overall Port to Remote Node Queue Time 936 Global Mirror Write I/O Rate 937 Global Mirror Overlapping Write Percentage 938 Global Mirror Overlapping Write I/O Rate 939 Peak Read Response Time 940 Peak Write Response Time 941 Global Mirror Secondary Write Lag 942 Table 7-22 lists SVC HBA ports metrics available for reports. 106 Reporting with TPCTOOL Table 7-22 SVC HBA ports metrics SVC - ctype:subsys_port 2 (HBA port) Metric Value Port Receive Data Rate 859 Port Send Data Rate 858 Port Send I/O Rate 852 Port Receive I/O Rate 853 Total Port I/O Rate 854 Total Port Data Rate 860 Port to Host Send I/O Rate 901 Port to Host Receive I/O Rate 902 Total Port to Host I/O Rate 903 Port to Disk Send I/O Rate 904 Port to Disk Receive I/O Rate 905 Total Port to Disk I/O Rate 906 Port to Local Node Send I/O Rate 907 Port to Local Node Receive I/O Rate 908 Total Port to Local Node I/O Rate 909 Port to Remote Node Send I/O Rate 910 Port to Remote Node Receive I/O Rate 911 Total Port to Remote Node I/O Rate 912 Port to Host Send Data Rate 913 Port to Host Receive Data Rate 914 Total Port to Host Data Rate 915 Port to Disk Send Data Rate 916 Port to Disk Receive Data Rate 917 Total Port to Disk Data Rate 918 Port to Local Node Send Data Rate 919 Port to Local Node Receive Data Rate 920 Total Port to Local Node Data Rate 921 Port to Remote Node Send Data Rate 922 Port to Remote Node Receive Data Rate 923 Total Port to Remote Node Data Rate 924 Table 7-23 lists SVC I/O group metrics available for reports. Chapter 7. Metrics per subsystem 107 Table 7-23 SVC I/O Group metrics SVC - ctype:svc_iogrp 5 108 (SVC I/O Group) Metric Value Total Data Rate 821 Read Response Time 822 Write Response Time 823 Port Receive Data Rate 859 Overall Response Time 824 Read Transfer Size 825 Write Transfer Size 826 Overall Transfer Size 827 Disk to Cache Transfer Rate 830 Cache to Disk Transfer Rate 831 Write-cache Delay Percentage 832 Write-cache Delay I/O Rate 833 Backend Read I/O Rate 835 Backend Write I/O Rate 836 Total Backend I/O Rate 837 Backend Read Data Rate 838 Backend Write Data Rate 839 Total Backend Data Rate 840 Backend Read Response Time 841 Backend Write Response Time 842 Overall Backend Response Time 843 Read Queue Time 844 Write Queue Time 845 Overall Queue Time 846 Backend Read Transfer Size 847 Backend Write Transfer Size 848 Overall Backend Transfer Size 849 Read I/O Rate (overall) 803 Write I/O Rate (overall) 806 Total I/O Rate (overall) 809 Read Cache Hits Percentage (overall) 812 Write Cache Hits Percentage (overall) 815 Reporting with TPCTOOL SVC - ctype:svc_iogrp 5 (SVC I/O Group) Total Cache Hits Percentage (overall) 818 Port Send Data Rate 858 Read Data Rate 819 Write Data Rate 820 Port Send I/O Rate 852 Port Receive I/O Rate 853 Total Port I/O Rate 854 Total Port Data Rate 860 Readahead Percentage of Cache Hits 890 Dirty Write Percentage of Cache Hits 891 Write Cache Overflow Percentage 894 Write Cache Overflow I/O Rate 895 Write Cache Flush-through Percentage 896 Write Cache Flush-through I/O Rate 897 Write Cache Write-through Percentage 898 Write Cache Write-through I/O Rate 899 CPU Utilization Percentage 900 Port to Host Send I/O Rate 901 Port to Host Receive I/O Rate 902 Total Port to Host I/O Rate 903 Port to Disk Send I/O Rate 904 Port to Disk Receive I/O Rate 905 Total Port to Disk I/O Rate 906 Port to Local Node Send I/O Rate 907 Port to Local Node Receive I/O Rate 908 Total Port to Local Node I/O Rate 909 Port to Remote Node Send I/O Rate 910 Port to Remote Node Receive I/O Rate 911 Total Port to Remote Node I/O Rate 912 Port to Host Send Data Rate 913 Port to Host Receive Data Rate 914 Total Port to Host Data Rate 915 Port to Disk Send Data Rate 916 Port to Disk Receive Data Rate 917 Chapter 7. Metrics per subsystem 109 SVC - ctype:svc_iogrp 5 (SVC I/O Group) Total Port to Disk Data Rate 918 Port to Local Node Send Data Rate 919 Port to Local Node Receive Data Rate 920 Total Port to Local Node Data Rate 921 Port to Remote Node Send Data Rate 922 Port to Remote Node Receive Data Rate 923 Total Port to Remote Node Data Rate 924 Port to Local Node Send Response Time 925 Port to Local Node Receive Response Time 926 Overall Port to Local Node Response Time 927 Port to Local Node Send Queue Time 928 Port to Local Node Receive Queue Time 929 Overall Port to Local Node Queue Time 930 Port to Remote Node Send Response Time 931 Port to Remote Node Receive Response Time 932 Overall Port to Remote Node Response Time 933 Port to Remote Node Send Queue Time 934 Port to Remote Node Receive Queue Time 935 Overall Port to Remote Node Queue Time 936 Global Mirror Write I/O Rate 937 Global Mirror Overlapping Write Percentage 938 Global Mirror Overlapping Write I/O Rate 939 Peak Read Response Time 940 Peak Write Response Time 941 Global Mirror Secondary Write Lag 942 Table 7-24 lists the SVC Managed Disk Group metrics available for reports. Table 7-24 SVC Managed Disk Group metrics SVC - ctype:svc_mdgrp 7 110 (SVC Managed Disk Group) Metric Value Total Data Rate 821 Read Response Time 822 Write Response Time 823 Overall Response Time 824 Reporting with TPCTOOL SVC - ctype:svc_mdgrp 7 (SVC Managed Disk Group) Read Transfer Size 825 Write Transfer Size 826 Overall Transfer Size 827 Backend Read I/O Rate 835 Backend Write I/O Rate 836 Total Backend I/O Rate 837 Backend Read Data Rate 838 Backend Write Data Rate 839 Total Backend Data Rate 840 Backend Read Response Time 841 Backend Write Response Time 842 Overall Backend Response Time 843 Read Queue Time 844 Write Queue Time 845 Overall Queue Time 846 Backend Read Transfer Size 847 Backend Write Transfer Size 848 Overall Backend Transfer Size 849 Read I/O Rate (overall) 803 Write I/O Rate (overall) 806 Total I/O Rate (overall) 809 Read Data Rate 819 Write Data Rate 820 Table 7-25 lists SVC Managed Disk metrics available for reports. Table 7-25 SVC Managed Disk metrics SVC - ctype:svc_mdisk 11 (SVC Managed Disk) Metric Value Backend Read I/O Rate 835 Backend Write I/O Rate 836 Total Backend I/O Rate 837 Backend Read Data Rate 838 Backend Write Data Rate 839 Total Backend Data Rate 840 Chapter 7. Metrics per subsystem 111 SVC - ctype:svc_mdisk 11 (SVC Managed Disk) Backend Read Response Time 841 Backend Write Response Time 842 Overall Backend Response Time 843 Read Queue Time 844 Write Queue Time 845 Overall Queue Time 846 Backend Read Transfer Size 847 Backend Write Transfer Size 848 Overall Backend Transfer Size 849 Table 7-26 SVC volume component metrics available for reports. Table 7-26 SVC volume component metrics SVC - ctype:vol 112 12 (VolumeComponent) Metric Value Total Data Rate 821 Read Response Time 822 Write Response Time 823 Overall Response Time 824 Read Transfer Size 825 Write Transfer Size 826 Overall Transfer Size 827 Disk to Cache Transfer Rate 830 Cache to Disk Transfer Rate 831 Write-cache Delay Percentage 832 Write-cache Delay I/O Rate 833 Read I/O Rate (overall) 803 Write I/O Rate (overall) 806 Total I/O Rate (overall) 809 Read Cache Hits Percentage (overall) 812 Write Cache Hits Percentage (overall) 815 Total Cache Hits Percentage (overall) 818 Read Data Rate 819 Write Data Rate 820 Readahead Percentage of Cache Hits 890 Reporting with TPCTOOL SVC - ctype:vol 12 (VolumeComponent) Dirty Write Percentage of Cache Hits 891 Write Cache Overflow Percentage 894 Write Cache Overflow I/O Rate 895 Write Cache Flush-through Percentage 896 Write Cache Flush-through I/O Rate 897 Write Cache Write-through Percentage 898 Write Cache Write-through I/O Rate 899 Global Mirror Write I/O Rate 937 Global Mirror Overlapping Write Percentage 938 Global Mirror Overlapping Write I/O Rate 939 Peak Read Response Time 940 Peak Write Response Time 941 Global Mirror Secondary Write Lag 942 Table 7-27 lists SVC node metrics available for reports. Table 7-27 SVC node metrics SVC - ctype:svc_node 15 SVC Node Metric Value Total Data Rate 821 Port Send Data Rate 858 Read Response Time 822 Write Response Time 823 Overall Response Time 824 Read Transfer Size 825 Write Transfer Size 826 Overall Transfer Size 827 Disk to Cache Transfer Rate 830 Cache to Disk Transfer Rate 831 Write-cache Delay Percentage 832 Write-cache Delay I/O Rate 833 Backend Read I/O Rate 835 Backend Write I/O Rate 836 Total Backend I/O Rate 837 Chapter 7. Metrics per subsystem 113 SVC - ctype:svc_node 114 15 SVC Node Backend Read Data Rate 838 Backend Write Data Rate 839 Total Backend Data Rate 840 Backend Read Response Time 841 Backend Write Response Time 842 Overall Backend Response Time 843 Read Queue Time 844 Write Queue Time 845 Overall Queue Time 846 Backend Read Transfer Size 847 Backend Write Transfer Size 848 Overall Backend Transfer Size 849 Read I/O Rate (overall) 803 Write I/O Rate (overall) 806 Total I/O Rate (overall) 809 Read Cache Hits Percentage (overall) 812 Write Cache Hits Percentage (overall) 815 Total Cache Hits Percentage (overall) 818 Read Data Rate 819 Write Data Rate 820 Port Send I/O Rate 852 Port Receive I/O Rate 853 Total Port I/O Rate 854 Port Receive Data Rate 859 Total Port Data Rate 860 Readahead Percentage of Cache Hits 890 Dirty Write Percentage of Cache Hits 891 Write Cache Overflow Percentage 894 Write Cache Overflow I/O Rate 895 Write Cache Flush-through Percentage 896 Write Cache Flush-through I/O Rate 897 Write Cache Write-through Percentage 898 Write Cache Write-through I/O Rate 899 CPU Utilization Percentage 900 Reporting with TPCTOOL SVC - ctype:svc_node 15 SVC Node Port to Host Send I/O Rate 901 Port to Host Receive I/O Rate 902 Total Port to Host I/O Rate 903 Port to Disk Send I/O Rate 904 Port to Disk Receive I/O Rate 905 Total Port to Disk I/O Rate 906 Port to Local Node Send I/O Rate 907 Port to Local Node Receive I/O Rate 908 Total Port to Local Node I/O Rate 909 Port to Remote Node Send I/O Rate 910 Port to Remote Node Receive I/O Rate 911 Total Port to Remote Node I/O Rate 912 Port to Host Send Data Rate 913 Port to Host Receive Data Rate 914 Total Port to Host Data Rate 915 Port to Disk Send Data Rate 916 Port to Disk Receive Data Rate 917 Total Port to Disk Data Rate 918 Port to Local Node Send Data Rate 919 Port to Local Node Receive Data Rate 920 Total Port to Local Node Data Rate 921 Port to Remote Node Send Data Rate 922 Port to Remote Node Receive Data Rate 923 Total Port to Remote Node Data Rate 924 Port to Local Node Send Response Time 925 Port to Local Node Receive Response Time 926 Overall Port to Local Node Response Time 927 Port to Local Node Send Queue Time 928 Port to Local Node Receive Queue Time 929 Overall Port to Local Node Queue Time 930 Port to Remote Node Send Response Time 931 Port to Remote Node Receive Response Time 932 Overall Port to Remote Node Response Time 933 Port to Remote Node Send Queue Time 934 Chapter 7. Metrics per subsystem 115 SVC - ctype:svc_node 15 SVC Node Port to Remote Node Receive Queue Time 935 Overall Port to Remote Node Queue Time 936 Global Mirror Write I/O Rate 937 Global Mirror Overlapping Write Percentage 938 Global Mirror Overlapping Write I/O Rate 939 Peak Read Response Time 940 Peak Write Response Time 941 Global Mirror Secondary Write Lag 942 7.5 Metrics for switch fabric Table 7-28 lists the switch fabric component metrics available for reports. Table 7-28 Switch fabric component metrics Component type available switch 13 Switch switch_port 14 Switch Port Table 7-29 lists the switch metrics available for reports. Table 7-29 Switch metrics Switch FABRIC - ctype: switch 116 13 (Switch) Metric Value Port Send Data Rate 858 CRC Error Rate 877 Link Failure Rate 874 Loss of Sync Rate 875 Port Send Packet Rate 855 Port Receive Packet Rate 856 Total Port Packet Rate 857 Port Receive Data Rate 859 Total Port Data Rate 860 Port Peak Send Data Rate 861 Port Peak Receive Data Rate 862 Port Send Packet Size 869 Port Receive Packet Size 870 Overall Port Packet Size 871 Reporting with TPCTOOL Switch FABRIC - ctype: switch 13 (Switch) Error Frame Rate 872 Dumped Frame Rate 873 Loss of Signal Rate 876 Short Frame Rate 878 Long Frame Rate 879 Encoding Disparity Error Rate 880 Discarded Class3 Frame Rate 881 F-BSY Frame Rate 882 F-RJT Frame Rate 883 Table 7-30 lists switch port metrics available for reports. Table 7-30 Switch port metrics Switch FABRIC - ctype: switch_port 14 (Switch Port) Metric Value Port Send Data Rate 858 Link Failure Rate 874 CRC Error Rate 877 Port Send Packet Rate 855 Port Receive Packet Rate 856 Total Port Packet Rate 857 Port Receive Data Rate 859 Total Port Data Rate 860 Port Peak Send Data Rate 861 Port Peak Receive Data Rate 862 Port Send Packet Size 869 Port Receive Packet Size 870 Overall Port Packet Size 871 Error Frame Rate 872 Dumped Frame Rate 873 Loss of Sync Rate 875 Loss of Signal Rate 876 Short Frame Rate 878 Long Frame Rate 879 Chapter 7. Metrics per subsystem 117 Switch FABRIC - ctype: switch_port 14 (Switch Port) 118 Encoding Disparity Error Rate 880 Discarded Class3 Frame Rate 881 F-BSY Frame Rate 882 F-RJT Frame Rate 883 Reporting with TPCTOOL Related publications The publications listed in this section are considered particularly suitable for a more detailed discussion of the topics covered in this Redpaper. IBM Redbooks For information about ordering these publications, see “How to get IBM Redbooks” on page 119. Note that some of the documents referenced here may be available in softcopy only. TotalStorage Productivity Center Advanced Topics, SG24-7348 Monitoring Your Storage Subsystems with TotalStorage Productivity Center, SG24-7364 How to get IBM Redbooks You can search for, view, or download Redbooks, Redpapers, Hints and Tips, draft publications and Additional materials, as well as order hardcopy Redbooks or CD-ROMs, at this Web site: ibm.com/redbooks Help from IBM IBM Support and downloads ibm.com/support IBM Global Services ibm.com/services © Copyright IBM Corp. 2007. All rights reserved. 119 120 Reporting with TPCTOOL Index Numerics 15K RPM DDMs 35 A AgentCLI 2 B backend IO operations 44 Backend IO rate 47 backend response time 45 C command mode 7 config files 6 creating reports 66 CSV file export 48 H Host Authentication Password 4 I import into template 77 interactive command mode 7 L large reads 32 Line graph 80 lscomp command 22 lsdev command 2, 17, 22 lsmetrics command 22, 54 lstime command 22 lstype command 22 M Multiple / Script command mode 8 D DS array metrics 101 DS component type metrics 95 DS controller metrics 98 DS device adapter metrics 99 DS HBA port metrics 98 DS rank metrics 100 DS subsystem metrics 96 DS volume component metrics 102 DS4000 HBA port metrics 86 DS4000 report metrics 86 DS4000 subsystem metrics 86 DS4000 volume metrics 87 E ESS array metrics 93 ESS component type metrics 87 ESS controller metrics 90 ESS device adapter metrics 91 ESS HBA port metrics 89 ESS subsystem metrics 87 ESS volume component metrics 94 exported data 66 F frontend I/O operations 44 G getrpt command 40, 54, 59, 65 globally-unique identifier 17 graphs 79 GUID 17 © Copyright IBM Corp. 2007. All rights reserved. N NVS Full Percentage 41 O OLTP applications 35 output to text file 24 Overall Backend Response time threshold 47 P perfcli 2 performance graph 83 performance metric 10 performance metrics rules of thumb 26 performance reports prerequisite tasks 16 performance snapshots 50 policy setting 50 Port Data Rate 51 Port Response Time 51 R RAID ranks 34 Random Read I/O 33 Read Cache Hit Percentage 41 Read Hit Percentages 32 Redbooks Web site 119 Contact us viii report graphs 79 report macro 73 report template 77 report timestamp considerations 72 121 reports CLI interface 14 response time chart 49 response time evaluation 39 response time factors 27 response time metrics 30 response times 29 rule of thumb rank 35 S scatter plots 80 SCRMCP 2 single-shot command mode 7 small block reads 32 small block writes 32 specifying a report delimiter 68 SVC component type metrics 103 SVC HBA port metrics 106 SVC I/O group metric 107 SVC Managed Disk Group metrics 110 SVC Managed Disk metrics 111 SVC metrics 58 SVC node metrics 113 SVC subsystem metrics 104 SVC volume component metrics 112 switch fabric component metrics 116 switch metrics 116 switch port metrics 117 switch ports reports 62 syntax to a file 8 T text file output 24 throughput measurements 44 throughput metrics 30, 38 TOTAL Backend I/O rate threshold 47 Total Cache Hit percentage 41 TOTAL I/O rate threshold 47 TPCCLI.CONF file 4, 6 TPCTOOL CLI 48 TPCTOOL overview 2 TPCTOOL report metrics 85 122 Reporting with TPCTOOL Back cover ® Reporting with TPCTOOL Redpaper Learn the reporting capabilities of TPCTOOL Create customized reports Evaluate report data TPCTOOL is a command line interface (CLI) based program which interacts with the TotalStorage Productivity Center Device server and it lets you create graphs and charts with multiple metrics, with different unit types, and for multiple entities (for example, subsystems, volumes, controller, and arrays). Commands are entered as lines of text and output can be received as text. INTERNATIONAL TECHNICAL SUPPORT ORGANIZATION This IBM Redpaper gives you an overview of the function of TPCTOOL and shows you how to use it to generate reports based on your TotalStorage Productivity Center repository data. BUILDING TECHNICAL INFORMATION BASED ON PRACTICAL EXPERIENCE IBM Redbooks are developed by the IBM International Technical Support Organization. Experts from IBM, Customers and Partners from around the world create timely technical information based on realistic scenarios. Specific recommendations are provided to help you implement IT solutions more effectively in your environment. For more information: ibm.com/redbooks REDP-4230-00