HP Matrix Operating Environment Software User Manual

HP Insight Capacity Advisor 6.2 User Guide

HP Part Number: T8670-90019 Published: November 2010 Edition: 2

Confidential computersoftware. Valid license from HP required for possession, use or copying. Consistent with FAR 12.211 and 12.212, Commercial

Computer Software, Computer Software Documentation, and Technical Data for Commercial Items are licensed to the U.S. Government under

vendor's standard commercial license.

The informationcontained hereinis subject to change without notice. Theonly warranties for HP productsand servicesare setforth in the express

warranty statements accompanying such products and services. Nothing herein should be construed as constituting an additional warranty. HP

shall not be liable for technical or editorial errors or omissions contained herein.

Acknowledgments

HP-UX Release 10.20 and later and HP-UX Release 11.00 and later (in both 32 and 64-bit configurations) on all HP 9000 computers are Open

Group UNIX 95 branded products.

UNIX is a registered trademark of The Open Group.

Intel and Itanium are trademarks or registered trademarks of Intel Corporation or its subsidiaries in the United States and other countries.

Microsoft and Windows are U.S. registered trademarks of Microsoft Corporation.

Revision history

Number

EditionDocument Part

Publication DateSupported Operating SystemsSoftware

Version

November 2010See the HP Insight Software 6.2 Support Matrix.6.2.02T8670-90019

October 2010See the HP Insight Software 6.2 Support Matrix.6.2.01T8670-90019

1 Introduction...................................................................................................................11

Setup considerations.............................................................................................................................12

Licensing requirements...................................................................................................................12

Installation.......................................................................................................................................12

Collection methods..........................................................................................................................12

Disk space requirements..................................................................................................................12

Credentials.......................................................................................................................................13

Dependencies...................................................................................................................................13

Upgrades and reinstallation.................................................................................................................13

Upgrading agents on HP-UX or OpenVMS managed systems......................................................13

Navigating within HP SIM, Virtualization Manager, and Capacity Advisor......................................13

2 Features.........................................................................................................................15

Data collection.......................................................................................................................................16

Data collection infrastructure..........................................................................................................16

Supported collection configurations...............................................................................................17

Comparison of agentless and UP data collection............................................................................18

HP PMP data differs from Utilization Provider data......................................................................19

Differences in memory data ......................................................................................................19

Differences in network data ......................................................................................................19

Related topics........................................................................................................................19

Experimentation....................................................................................................................................20

Sizing for service level objectives..........................................................................................................21

Modeling considerations......................................................................................................................22

Capacity Advisor commands...............................................................................................................22

3 Key Capacity Advisor concepts..................................................................................23

Capacity planning goals.......................................................................................................................23

Quality of service ............................................................................................................................23

Utilization monitor, calculator, and simulator.....................................................................................23

Data handling for virtual machines.................................................................................................23

Data normalization..........................................................................................................................24

Resources for which Capacity Advisor collects data............................................................................24

Upper bounds..................................................................................................................................24

Measuring and analyzing resource utilization.....................................................................................25

Peaks and sums................................................................................................................................25

Sampling interval.............................................................................................................................25

Headroom........................................................................................................................................25

Headroom star rating.................................................................................................................26

Interpreting the headroom star rating..................................................................................26

Missing or invalid data....................................................................................................................27

The affect of invalid data on HP Smart Solver solution finding................................................27

Utilization limits .............................................................................................................................27

Specifying utilization limits.......................................................................................................28

Sustained time limits.............................................................................................................28

Percentage of time limits.......................................................................................................28

Understanding utilization limit messages.................................................................................29

Percentage of allocation........................................................................................................29

With sustained limits............................................................................................................29

Table of Contents 3

With percentage of time limits..............................................................................................29

Scope of utilization limits...........................................................................................................29

Adjusting power...................................................................................................................................30

Power cap data.................................................................................................................................30

Automating solution generation: HP Smart Solver..............................................................................31

HP Smart Solver: Types of solutions...............................................................................................31

Determining trends in Capacity Advisor.............................................................................................31

Aggregation of points in business interval bins..............................................................................32

Choosing an appropriate business interval...............................................................................32

Exclusion of data points..................................................................................................................32

Factors that affect data validity..................................................................................................32

Linear regression.............................................................................................................................33

Error analysis...................................................................................................................................33

Forecasting growth...............................................................................................................................33

The forecast model hierarchy..........................................................................................................33

Forecast model attributes................................................................................................................34

4 Planning with Capacity Advisor.................................................................................35

Getting ready........................................................................................................................................35

Task: Understand current resource usage............................................................................................35

Task: Plan server consolidation.............................................................................................................36

Understanding the consolidation task............................................................................................36

Example consolidation: Stacking applications on an existing server........................................36

Step 1: Determine which systems to consolidate..................................................................37

Step 2: Create a scenario.......................................................................................................37

Step 3: Edit the scenario........................................................................................................38

Automating the consolidation task.................................................................................................46

Example consolidation: Automating stacking on a “what-if” server........................................47

Step 1: Determine which systems to consolidate..................................................................47

Step 2: Create a scenario.......................................................................................................47

Step 3: Edit the scenario........................................................................................................48

Task: Estimating the effect of adding or moving processors................................................................51

Task: Determining where to put a workload using automated solution finding.................................52

Example: Determining where to put a workload............................................................................52

5 Procedures....................................................................................................................55

Accessing Capacity Advisor.................................................................................................................55

Gathering data for Capacity Advisor...................................................................................................55

Impact of data collection on managed system performance...........................................................56

Data collection options....................................................................................................................56

First data collection (or the automated nightly collection).............................................................58

Interpreting task results...................................................................................................................59

Error notification on the standard out tab.................................................................................59

Scheduling a data collection............................................................................................................60

Modifying a collection schedule......................................................................................................61

Removing a collection schedule......................................................................................................61

Updating collected data...................................................................................................................61

Updating collected data on all systems......................................................................................62

Updating collected data on selected systems.............................................................................62

Updating data displayed in a profile viewer.............................................................................62

Collecting data without using an agent..........................................................................................62

Listing systems currently in the configuration file....................................................................62

Setting advanced options in the agentless data collection file...................................................63

4 Table of Contents

Data collection and the HP Insight Capacity Advisor Consolidation software..............................63

Using Capacity Advisor with consolidation software server data............................................64

Viewing the licensed servers in HP SIM....................................................................................64

Importing data for use in Capacity Advisor...................................................................................64

From the Optimize menu — Import HP OVPA data.................................................................64

From the Optimize menu — Import HP PMP data...................................................................65

From the command line.............................................................................................................66

Ascertaining the data collection availability for a set of servers................................................66

Possible sources of discrepancies in collected data.........................................................................66

No data collected for a managed node......................................................................................66

Data timestamp appears to be incorrect....................................................................................66

Dynamic memory in HP Integrity virtual machines.................................................................67

Producing graphs and reports..............................................................................................................67

Using the Profile Viewer..................................................................................................................67

Using the report wizard..................................................................................................................71

The report wizard.......................................................................................................................71

Creating an historic utilization report........................................................................................72

Creating a scenario utilization report........................................................................................73

Using the report wizard to create a scenario comparison.........................................................74

An example scenario comparison report..............................................................................74

Calculating a virtualization consolidation ratio...................................................................75

Creating a consolidation candidates report...............................................................................76

Creating a cost allocation report................................................................................................76

Creating a peak summary report...............................................................................................76

Creating a population report.....................................................................................................77

Creating a trend report...............................................................................................................77

Creating a power report.............................................................................................................78

Setting utilization limits........................................................................................................................78

Setting global utilization limits.......................................................................................................79

Adding a utilization limit...........................................................................................................79

Removing a utilization limit......................................................................................................79

Setting workload utilization limits..................................................................................................80

Adding a utilization limit...........................................................................................................80

Enabling or disabling the limit...................................................................................................80

Removing a utilization limit......................................................................................................80

Setting scenario-wide utilization limits...........................................................................................80

Adding a utilization limit...........................................................................................................80

Enabling or disabling the limit...................................................................................................81

Removing a utilization limit......................................................................................................81

Setting scenario workload utilization limits....................................................................................81

Adding a utilization limit...........................................................................................................81

Enabling or disabling the limit...................................................................................................81

Removing a utilization limit......................................................................................................82

Forecasting utilization..........................................................................................................................82

Defining forecast models.................................................................................................................82

Accessing the Global Forecast Model........................................................................................82

Defining the Global Forecast Model.....................................................................................83

Accessing the forecast model for a workload or system............................................................83

Defining the forecast model for a workload or system........................................................84

Accessing the forecast model for a scenario..............................................................................84

Accessing the forecast model for a workload within a scenario................................................85

Defining a forecast model..........................................................................................................85

Disabling a forecast model.........................................................................................................85

Enabling a forecast model..........................................................................................................85

Generating forecasts........................................................................................................................85

Table of Contents 5

Viewing forecast data in a profile viewer..................................................................................86

Viewing forecast data in a utilization report.............................................................................86

Working with scenarios........................................................................................................................86

Creating a planning scenario...........................................................................................................86

Editing a scenario............................................................................................................................88

Controlling the data display............................................................................................................88

Change the meter style selection................................................................................................88

Change the data collection period by setting the data range.....................................................88

Change the meter representation...............................................................................................90

Copying a scenario..........................................................................................................................90

Renaming a scenario........................................................................................................................90

Undo/edit/review applied what-if actions in a scenario.................................................................91

To view applied what-if actions.................................................................................................91

To edit applied what-if actions...................................................................................................91

Modifying resource utilization values..................................................................................92

Deleting a scenario..........................................................................................................................92

Working with systems..........................................................................................................................93

Creating a system............................................................................................................................93

Adding an existing system..............................................................................................................94

Editing a system..............................................................................................................................94

Changing VMs to be servers............................................................................................................94

Change VMs to server................................................................................................................95

Changing servers to be VMs (manually).........................................................................................95

Change servers to be VMs..........................................................................................................95

Assigning VM hosts to a VMware DRS cluster in a scenario..........................................................96

Initially converting VM hosts to simulate a VMware DRS cluster............................................96

Adding VM hosts to a converted cluster...................................................................................97

Removing a VM host from a VMware DRS cluster...................................................................97

Removing a VMware DRS cluster from a scenario....................................................................97

Moving a virtual machine...............................................................................................................98

Removing a system..........................................................................................................................99

Editing network and disk I/O capacity...........................................................................................99

Getting there...............................................................................................................................99

Setting the upper bound values.................................................................................................99

Working with workloads....................................................................................................................100

Introduction...................................................................................................................................100

Creating a workload......................................................................................................................100

Editing a workload........................................................................................................................101

Moving a workload........................................................................................................................102

Move considerations.................................................................................................................102

Parking a workload........................................................................................................................104

Deleting a workload......................................................................................................................104

Working with power...........................................................................................................................104

Calibrating power within Virtualization Manager........................................................................105

Calibrating a single system......................................................................................................105

Calibrating multiple systems at once.......................................................................................107

Calibrating power within a scenario.............................................................................................107

Getting there.............................................................................................................................108

Calibrating a single system......................................................................................................108

Calibrating multiple systems at once.......................................................................................109

Automating time-consuming simulations..........................................................................................109

Automated solution finding: System consolidation to VMs.........................................................109

Begin with a scenario................................................................................................................110

Choose the systems to consolidate...........................................................................................110

Step 1 of 3: Define the destination system(s) and attributes....................................................110

6 Table of Contents

Step 2 of 3: Set level of effort for Smart Solver calculation.......................................................111

Step 3 of 3: View Smart Solver solutions..................................................................................111

Results: Automated consolidation to VMs.........................................................................112

Automated solution finding: Load balance of servers or VM hosts .............................................113

Begin with a scenario................................................................................................................113

Choose the VM hosts or servers to load-balance.....................................................................113

Step 1 of 3: Define the solution constraints..............................................................................114

Step 2 of 3: Set level of effort for Smart Solver calculation.......................................................114

Step 3 of 3: View Smart Solver solutions..................................................................................114

Results: Automated load balancing of servers or VM hosts...............................................114

Automated solution finding: Workload stacking..........................................................................115

Begin with a scenario................................................................................................................116

Choose the workloads to stack.................................................................................................116

Step 1 of 3: Stack the specified workloads onto servers...........................................................116

Step 2 of 3: Set level of effort for Smart Solver calculation.......................................................117

Step 3 of 3: View Smart Solver solutions..................................................................................117

Results: Automated workload stacking..............................................................................117

Getting more detail........................................................................................................................118

6 Using Capacity Advisor with HP Serviceguard......................................................119

Using Serviceguard to migrate HP Integrity VM guests....................................................................119

Data collection in a Serviceguard environment..................................................................................120

7 Support and other resources....................................................................................121

Contacting HP.....................................................................................................................................121

New and changed information in this edition....................................................................................121

New and changed information in Edition 1..................................................................................121

Related information............................................................................................................................121

Related technical papers................................................................................................................122

A Calculation assistance..............................................................................................123

Cost calculation...................................................................................................................................123

Cost per kilowatt-hour...................................................................................................................123

Cooling calculation.............................................................................................................................123

Cooling multiplier.........................................................................................................................123

Adjusting for platform changes..........................................................................................................123

Memory multiplier........................................................................................................................124

Adjusting for change in a workload...................................................................................................124

CPU workload multiplier..............................................................................................................124

Memory workload multiplier........................................................................................................125

Network I/O workload multiplier.................................................................................................125

Disk I/O workload multiplier........................................................................................................125

Determining estimated utilization assumptions for a workload..................................................125

Adjusting for virtualization changes..................................................................................................127

CPU Virtualization Overhead %....................................................................................................127

CPU Virtualization Overhead %..............................................................................................128

Hypervisor memory overhead......................................................................................................128

Hypervisor memory overhead.................................................................................................128

Doing the math for hypervisor memory overhead..................................................................128

HP Virtual Machine............................................................................................................128

VMware ESX 3.....................................................................................................................128

VMware vSphere (ESX 4)....................................................................................................129

Microsoft Hyper-V..............................................................................................................129

Table of Contents 7

Power calibration calculations............................................................................................................129

Determining idle/max values for non-Blade servers ....................................................................129

Determining idle/max values for Blade servers............................................................................130

P-class HP Blades and HP Blade system sizer:........................................................................130

C-class HP Blades and HP Blade system sizer:........................................................................131

B Command reference..................................................................................................133

capagentlesscfg (1M)...........................................................................................................................134

capcollect (1M)....................................................................................................................................136

capcustombenchmark(1M).................................................................................................................139

capcustombenchmark (4)....................................................................................................................141

capovpaextract (1M)............................................................................................................................142

capprofile (1M)....................................................................................................................................144

capprofile (4).......................................................................................................................................148

cappmpextract ....................................................................................................................................151

C Units and terminology...............................................................................................153

D Snapshots of Capacity Advisor reports...................................................................157

Consolidation candidate report..........................................................................................................157

Scenario comparison report................................................................................................................157

E Graphed data in Capacity Advisor reports............................................................161

F Example of the Undo/Edit/View Applied What-If Actions screen.......................163

G Troubleshooting in Capacity Advisor ....................................................................165

When there is no connection...............................................................................................................165

Data is not being collected..................................................................................................................166

Data may appear to be old when it is not...........................................................................................167

Data seems to be incorrect or lost.......................................................................................................168

Data cannot be imported.....................................................................................................................170

Out-of-memory errors in the Java heap..............................................................................................170

When working with reports................................................................................................................171

When working in the scenario editor..................................................................................................171

When working with power.................................................................................................................172

Workload and system relationships may not match when comparing Capacity Advisor and

Virtualization Manager information...................................................................................................173

Miscellaneous user interface issues....................................................................................................173

H Capacity Advisor messages.....................................................................................175

Command error messages..................................................................................................................175

Command warning messages.............................................................................................................176

HP Smart Solver error messages.........................................................................................................180

When available systems lack sufficient capacity...........................................................................180

When the threshold for invalid data is exceeded..........................................................................181

HP Smart Solver informational messages...........................................................................................182

Glossary.........................................................................................................................183

8 Table of Contents

Index...............................................................................................................................191

Table of Contents 9

1 Introduction

This document describes how to use Capacity Advisor to plan for changes in resource usage in a virtualized server environment. Checklists and examples help you with common planning tasks. This document is for technical professionals working in data center operations, administration, and planning. Some understanding of system administration is assumed.

HP Insight Capacity Advisor software is a program that allows you to monitor and evaluate system and workload utilization of CPU cores, memory, network and disk I/O, and power. With this information, you can load your systems to make best use of the available resources.

You can monitor and evaluate one or more systems that are connected in a cluster configuration or to a network. A single system can include multi-core or hyper-threaded processors.

Capacity Advisor helps you evaluate system consolidations, load balancing, changing system attributes, and varying workloads to decide how to move workloads to improve utilization. The quantitative results from Capacity Advisor can aid the planner in estimating future system workloads and in planning for changes to system configurations. With Capacity Advisor, you can perform the following tasks within an easy-to-navigate, clearly notated graphical user interface:

• Collect utilization data on CPU cores, memory, network and disk I/O, and power.

• View historical resource utilization for whole-OS and monitored workloads on HP-UX and

OpenVMS systems andwhole-OS workload resourceutilization on Microsoft Windows and Linux systems.

• View historical workload resource utilization and aggregateutilization across the partitioning

continuum (see the HP Insight Software 6.2 Support Matrix for a complete list of supported platforms).

• Generate resource utilization reports.

• Plan workload or system changes, and assess impact on resource utilization.

• Assess resource utilization impact for proposed changes in workload location or size.

• Evaluate trends for forecasting resource needs.

Capacity Advisor can be used to simulate changes in system configuration, such as the following:

• Consolidating several systems into one system

• Re-sizing a system for an upgrade

• Re-sizing the demands on a system to reflect a forecast

• Replacing older, small to mid-sized systems with virtual machines

Capacity Advisor can use data collected over time to show the results of these configuration changes in many ways. Graphical views are available so you can see what the effects of the changes are over time. Tables are available that give the percentage of time and the degree to which the system is busy; this information is valuable in comparing resource utilization and

quality of service before and after a change. Other tables show how many minutes per month the

system is unacceptably busy–a measure valuable for both quality of service and for estimating TiCAP bills. Because Capacity Advisor works from data traces collected over time, it is much more accurate than using only peak data or average data in understanding your systems and the workloads they support.

The significant advantage in using Capacity Advisor, rather than ad hoc processes or guesswork, is that it provides a quantitative basis for examining the usage of current resources. Additionally, it provides the capability to try simulations (what-if scenarios) for moving workloads or other resources before you actually implement a move.

Capacity Advisor incorporates numerical values of several components in its analysis and modeling, including:

• Number of CPU cores

• CPU benchmark data

• Memory size

• Network I/O bandwidth

• Disk I/O bandwidth

• Power usage

• Platform multiplier for memory (see “Adjusting for platform changes” (page 123))

• Virtualization adjustments (see “Adjusting for virtualization changes” (page 127))

Setup considerations

To prepare to use Capacity Advisor, consider the following:

• Licensing requirements

• Installation

• Upgrade and reinstallation

• Credentials

• Dependencies

Licensing requirements

Capacity Advisoris installed when HP Insight Dynamics suite is installed. For specificinformation about trial and LTU (License To Use) licensing on supported platforms, see the HP Insight Dynamics

6.2 Getting Started Guide at http://www.hp.com/go/insightdynamics/docs. For added information on using the HP Insight Capacity Advisor Consolidation software license, see “Data collection

and the HP Insight Capacity Advisor Consolidation software” (page 63). Capacity Advisor and

Capacity Advisor Consolidation Software licenses cannot be used on the same system at the same time.

Installation

Capacity Advisoris installed with other Insight Dynamics components, so no separate installation is required. (See the Insight Dynamics installation guide at http://www.hp.com/go/

insightdynamics/docs appropriate for your operating system.) To use Capacity Advisor, you

must have a valid license.

Collection methods

On installation of HP Insight Dynamics suite, the Insight managed system setup wizard automatically configures the mechanism used to collect data from managed servers based on the type of operating system that the managed system is running. (See “Supported collection

configurations” (page 17).)

Some problems with data collection can be diagnosed with vseassist. See the HP Insight Dynamics 6.2 Getting Started Guide for additional information.

Disk space requirements

Plan to allot the following amounts of memory for each workload to collect and preserve a maximum of four years of data for use by Capacity Advisor:

• On a Microsoft Windows CMS: Allot 64.5 MB foreach workload on each Windows or Linux

managed node to support agentless data collection. Allot 34.7 MB for each workload on each HP-UX managed node.

• On an HP-UX CMS: Allot 93.1 MB for each workload on each Integrity Linux managed

node to support agentless data collection. Allot 34.7 MB for each workload on each Integrity Windows or HP-UX managed node.

12 Introduction

For additional information on disk space requirements, see the Insight Dynamics installation guide appropriate for your operating system at http://www.hp.com/go/insightdynamics/docs .

Credentials

To use Capacity Advisor, to collect data, and to run reports, you must have credentials (a valid user name and password) and appropriate toolbox authorization on systems where you plan to use Capacity Advisor. For specifics about setting up user authorizations, see the HP Insight Dynamics 6.2 Getting Started Guide at http://www.hp.com/go/insightdynamics/docs.

Dependencies

Data collection requires that credentials for each managed server be configured in HP SIM.

• WBEM/WMI credentials must be configured for all Windows servers, HP-UX servers, and

HP Integrity VMs. For agentless data collection on Windows systems, the WBEM/WMI credentials must be set to Administrator level. WBEM certificate authentication is supported for HP-UX servers.

• SSH credentials (user-based, host-based key, or by certificate) must be configured for all

Linux servers. Linux agentless data collection does not require root-level access. HP SIM Privilege elevation is supported on HP-UX and Linux servers using a single user account with sudo or powerbroker with no password required.

• The Windows or Linux managed node must be configured for agentless data collection by

running the Insight managed system setup wizard, or by running Collect Capacity Advisor Data... (capcollect) on the managed server. Administrator or root level permission on the CMS is required to configure agentless collection on the CMS.

• HP Insight Control virtual machine management (VMM) is required to collect data from

ProLiant/x86-based hypervisors and their VMs. Hyper-V VMs also require that WBEM/WMI credentials be configured for Windows VMs and SSH for Linux VMs.

Upgrades and reinstallation

When upgrading or reinstalling software used with Capacity Advisor, there are different considerations and procedures to follow specific to the operating systems underlying the CMS or running on the managed node. Refer to the appropriate installation guide for instructions. Install guides can be downloaded from http://www.hp.com/go/insightdynamics/docs.

Upgrading agents on HP-UX or OpenVMS managed systems

NOTE: This section does not apply to Microsoft Windows or to Linux systems, for which data

is collected by leveraging native instrumentation (agentless data collection).

When upgrading the Utilization Provider to the latest version on a managed node, any data collected since the last capcollect command execution is lost. To minimize data loss, run the capcollect command on the CMS to collect from the managed node just before upgrading.

Navigating within HP SIM, Virtualization Manager, and Capacity Advisor

Avoid using your browser Back and Forward buttons (except when navigating in the online help). When you use these buttons, the user interface cannot tell that you have changed to a different view. This can cause problems, including incorrectly displaying which objects are selected in the current view, losing information input into screen fields, and cancelling operations. Instead, use the navigation links, buttons, and tabs provided by the user interfaces within the browser screen area.

Upgrades and reinstallation 13

2 Features

Capacity Advisor lets you test configuration changes before actually making them in your data center, and helps you to manage existing resources to improve their utilization. For example, Capacity Advisor assists you in answering questions such as the following:

• Is there room on this system to support additional work?

• Will this workload fit on this system?

• Can these servers be consolidated as virtual machines on a single server?

• What might my resource demand be in six months or a year?

• What are the estimated power costs for cooling a real or simulated system configuration?

• What doesa comparison of several “what-if” scenarios reveal about the analyzed alternatives?

• What is the historic behavior of a system or workload?

Capacity Advisor can display data about the following system resources:

• CPU cores (number) and speed

• Memory

• Network I/O bandwidth

• Disk I/O bandwidth

• Power usage

Capacity Advisor enables you to collect resource data and create visualization graphs with the following views:

• Historical utilization and trends

• Data peaks

• Peak durations

• Forecasts

with the following utilization limits:

• amount of acceptable resource usage

• sustained time

• percentage of time

Capacity Advisor enables you to create scenarios for “what-if” planning and forecasting, performing tasks such as the following:

• For systems in scenarios:

— Create systems based on existing systems or with characteristics that you define — Remove systems — Edit system attributes — Turn systems into virtual machines — Move virtual machines

• For workloads in scenarios:

— Create workloads based on existing workloads or with characteristics that you define — Collect utilization data for workloads — Import existing workload profiles — Edit workload demands — Move workloads between systems — Park workloads — Delete workloads

• For scenarios themselves:

— View and undo/redo scenario changes — Use the HP Smart Solver technology to perform:

Automated server consolidation to virtual machines◦

◦ Automated load balancing of existing VM hosts or existing servers ◦ Automated workload stacking (consolidation)

— Create scenario comparison reports

Data collection

Data collection is at the heart of Capacity Advisor and a task is created during configuration to automatically collect utilization data from all resources licensed for Insight Dynamics and discovered by HP Systems Insight Manager software (HP SIM). (See “First data collection (or

the automated nightly collection)” (page 58).)

HP recommends that once you have become familiar with using Capacity Advisor, you should review this scheduled task to understand what it is doing and make sure that it fits your needs. This automated data collection task gathers historical data from all your systems so that the data is available when you need it for analysis.

Data collection infrastructure

In the data collection infrastructure, each component plays a role, as shown in Figure 2-1 “Capacity

Advisor Data Collection Infrastructure”. It all begins with data collection.

Figure 2-1 Capacity Advisor Data Collection Infrastructure

16 Features

The Agentless Data Collector Services run on the CMS and collect data continued use in the

The Data Collector gathers data from the agentless data files, the Utilization

Providers running on thefrom Microsoft Windows and Linux managed systems. Agentless data collected by the Agentless Data

managed nodes, and

from VMM.

You can import data (HP

OVPA and HP PMP) into Collector Service is made the historical database for

analysis.

Scenario data is stored automatically for

Scenario Editor.

Use the Scenario Editor to manipulate the collected data associated with workloads and systems.

Use a Profile Viewer to examine a graphical

available to the Capacity Advisor Data Collector.

The Utilization Provider enables the operation of the Capacity Advisor Data Collector. HP Insight Control virtual machine

Automatic data conversion occurs when importing data into the Insight Dynamics database. Historical data is stored for use in Capacity Advisor.

104

The Utilization Calculator uses historical utilization

display of the utilization of a specific resource for a specific workload or system.

149

Use the Report Generator to create reports for understanding utilization.

Use the web browser to work with Capacitydata and scales, sums, ormanagement (virtual

machine management) data is made available to the Capacity Advisor Data Collector.

Characteristics of

converts it to approximate actual or hypothetical system utilization.

Advisor and exploit its features.

managed systems are stored in the Insight Dynamics database and used when collecting utilization Data.

Collecting data for a workload with a HP Serviceguard package

You can associate a monitored workload with a Serviceguard package. This association allowsVirtualization Manager to show workloads moving from one system to another within minutes of a Serviceguard migration. It also allows Capacity Advisor to show an uninterrupted history of utilization data for a workload across migrations.

Associating monitored workloads and Serviceguard packages requires that Serviceguard and the corresponding version of the Utilization Provider have been installed on all member systems of the Serviceguard cluster. For specific version information, see the HP Insight Software 6.2

Support Matrix.

Supported collection configurations

Table 2-1 Collection methods by operating system on CMS and managed node types

Managed node software and/or OS

Integrity Virtual Machine (HP VM)

Collector for HP-UX CMSCollector for Microsoft Windows CMS

not supportedVMMVMM host or guest

not supportedVMM and agentless methodVMM Hyper-V VM

HP VMProvider and Utilization ProviderHP VM Provider and agentless methodMicrosoft Windows on HP

HP VM Provider and agentless methodHP VM Provider and agentless methodLinux on HP VM

HP VMProvider and Utilization ProviderHP VMProvider and Utilization ProviderHP-UX on HP VM

Utilization ProviderUtilization ProviderOpenVMS on HP VM

Utilization Provider (Integrity only)agentless methodstandalone Windows

agentless methodagentless methodstandalone Linux

Utilization ProviderUtilization Providerstandalone HP-UX

Utilization ProviderUtilization Providerstandalone OpenVMS

Data collection 17

1 Data is obtained from some managed system types using more than one collector. This is done where the data from

both collectors provide a broader or more accurate view than can be obtained from using only one collector on that

managed system type. For example, where bothnatively collecteddata (agentless) and data from the IntegrityVirtual

Machines Provider are used, the HP VM agent provides info about the VM that allows Capacity Advisor to obtain

more accurate CPU data for the VM, and the agentless method provides metrics other than CPU utilization.

NOTE: The supported collection configurations for a managed node are normally set up by

the Insight managed system setup wizard. If the capcollect command is run before the wizard is run on a managed node, and agentless data collection is supported for the managed node, the node will be configured for agentless collection at that time. Because no agentless data is yet available, however, the message:

Error: At the time data collection was run, system "system-name" was not configured for agentless data collection. It has been configured and data collection will begin shortly. No utilization data is available now. Try collecting again in 5 minutes.

will appear in the command output. Subsequent collections will retrieve collected data beginning at the time the capcollect command was run on the managed node.

Comparison of agentless and UP data collection

Table 2-2 (page 18) compares the operational characteristics of agentless data collection and that

of the Utilization Provider (UP) when used to collect data from systems licensed for Capacity Advisor.

• Agentless Data Collection. This method leverages native data collection already occurring

on the managed server(s).

• WBEM Utilization Provider. This method uses an agent to gather the data and provide it

to the CMS.

Table 2-2 A comparison of agentless data collection and use of Utilization WBEM Provider

Utilization Provider (UP)Agentless data collection

Equally useful in data quality and accuracy.

Available for HP-UX and OpenVMS systemsAvailable for Microsoft Windows, Linux, and non-HP

systems

Provides the basic set of utilization metrics for CPU, memory, network I/O, and disk I/O, plus additional metrics for reporting specific activities in the operating system, disk, and network: CPU queue length, page faults/second, disk I/Os per second, LAN packets/second, number of disks, disk space total, disk space used.

hosting or update required

Susceptible to network or CMS downtime as data is continuously collected by the CMS across the network from each managed system

Small additional load on the CMS CPU and memory resources incurred by the ongoing operation of the agentless data collection service

Provides the basic set of resource utilization metrics for CPU, memory, network I/O, and disk I/O.

Requires hosting and updating the UP agentLeverages existing data on thesystem; no additional agent

Not susceptible to downtime in the network or the CMS. Collected data is held for a maximum of 30 days by the UP for transfer to the database on the CMS

Slight additionalload on CMS CPU or memory resources incurred once per day

Slight additional load on managed system resources due to remote queries from the collector every 5 minutes

18 Features

Slight additionalload on managed system resources, due to local queries from the UP agent every 5 minutes, and storage/retrieval of historical metrics on the local file system

Table 2-2 A comparison of agentless data collection and use of Utilization WBEM Provider (continued)

Utilization Provider (UP)Agentless data collection

Slight additional load on network incurred once per daySmall additional load on the network incurred by the

ongoing operationof the agentless data collection service

Data collection starts within a few minutes of configuring the system using Insight managed system setup wizard

Data collection starts within 24 hours of configuring the system using Insight managed system setup wizard. However, if the UP is already running on the managed system, the data collected may cover an interval as long as 30 days previous to configuration inInsight Dynamics.

HP PMP data differs from Utilization Provider data

Differences in data imported from HP Performance Management Pack (HP PMP) affect how you should interpret resulting analysis and scenarios as compared to data collected from theUtilization Provider.

Differences in memory data

Capacity Advisor normally measures memory collected from the Utilization Provider, which returns the sum of memory allocated to the processes on the system.

HP PMP collection provides the total memory less the zeroed, free, and standby memory.

This means that HP PMP counts memory used by the operating system, whereas the Utilization Provider does not. The memory use reported by HP PMP is about 20% higher than memory use reported by the Utilization Provider.

Because Capacity Advisor assumes a different memory amount than HP PMP provides, Capacity Advisor may overestimate the memory needed when using HP PMP data to analyze stacking several application instances in a system running a single instance of an operating system.

While either the Utilization Provider data or the HP PMP data can be used to plan memory size in a new configuration, you should not use a mix of HP PMP and Utilization Provider data when looking for a trend in memory utilization.

Differences in network data

HP PMP does not report the correct network I/O when network interface cards (NICs) are being teamed for network fault tolerance (for example, by using HP Network Config utility).

That is, for a system with 2 NICs teamed as one single logical NIC, Utilization Provider reports the addition of TX and RX Mbps for the teamed NIC.

HP PMP adds the metrics from the 2 NICs, ignoring that they are working as a teamed NIC.

For a teamed configuration, the HP PMP network utilization curve will be always twice that of the Utilization Provider data.

If you know you have teamed NICs, or if you suspect your imported network data is twice as high as it should be, you will want to instruct Capacity Advisor to ignore the imported data. When logged in as a user authorized to use the Capacity Advisor toolbox on the CMS, you will be able to mark data as invalid. You can do this from a profile viewer, or by using the capprofile(1M) command.

Experimentation

You can better understand Capacity Advisor by experimenting — considering different configurations and workload placement trying them out in “what-if” scenarios. A scenario identifies the workload demand profile that creates your experimental simulations.

When you do workload analysis in Capacity Advisor, you view graphs and reports that represent CPU or memory utilization by time. For example, Figure 2-2 shows a graph of CPU utilization for a single system over a one-month period.

Figure 2-2 CPU utilization for managed system puny03v8

Peak (highest) value.

Similarly, Figure 2-3 shows CPU utilization for a second system over the same period.

Figure 2-3 CPU utilization for managed system puny03v7

Peak value.

Comparing these two graphs shows that workload peaks on the two systems do not occur simultaneously, nor do they require the samepercentage of the allocated CPU cores for processing. This suggests an opportunity to consider whether you can consolidate both systems together to satisfy the needs of the workloads, while reducing the number of CPU cores (originally each system is allocated 2 cores, for a total of 4 cores available to do work).

Figure 2-4 shows the result of using a Capacity Advisor “what-if” scenario to combine the

workloads onto one system.

20 Features

Figure 2-4 Combined “what-if” CPU utilization for puny03v8 and puny03v7

Peak value.

From the graph, it is evident that the peak of the combined workloads is under 2 CPU cores. Even with utilization limits in place, this system is unlikely to need 4 CPU cores to meet this workload demand.

Sizing for service level objectives

As part of planning configuration changes, you can also consider your service level objectives or quality of service goals.

The following utilization graph shows only one month of data; longer periods of data provide an increasingly accurate picture of resource usage.

Figure 2-5 CPU utilization with selected percentiles of utilization

In this example, an allocation of three CPU cores is assumed, rather than the four cores initially available in this exercise. Even this reduction may not provide the best fit, as the vast majority of the work (90%) is completed with less than .5 of one core, and 99% of the work is completed with one core.

For example, as shown in Figure 2-5 (page 21), CPU utilization has one peak at 1.7 CPU cores, with many lower peaks. If you configure your system to always meet the demand of this single

1.7-CPU peak, and you do not adjust the CPU allocation, a significant fraction of the CPU allocation in this example would go unused most of the time. Depending on your quality of service goals, you may decide that a different configuration can better use the resources available.

Sizing for service level objectives 21

Further experimentation in Capacity Advisor with resource allocations, consolidations, and utilization limits will help you arrive at the best fit for these workloads.

With Capacity Advisor's visualization and reporting tools, you can make a considered estimate of server resource utilization using different scenarios and easily refine allocations by tweaking values in the scenarios. Such estimates can help you minimize overspending for capacity you use rarely, and maximize utilization of your systems to ensure that your systems have the capacity needed at the level you require.

Modeling considerations

Capacity Advisormethodology makes certain assumptions inits data analysis. Theseassumptions include using an automatically determined CPU performance index, with normalization based on benchmark data or benchmark data in combination with CPU clock speed in special cases.

Memory is not normalized. Scale factors can be set on the Capacity Advisor Edit Workload and Move Workload screens. See“Adjusting for platform changes” (page 123) for more information.

On the Edit Scenario screen, you can select either the System or Workload tab to manipulate the planning scenario and specify which metric representation to view: average, 90th percentile, peak or max 15-min (maximum 15-minute sustained). The Meter Style can be absolute numbers or in percentages. By changing the metrics to view, you can explore multiple options for system and workload deployment. See “ Controlling the data display” (page 88) for more information.

Capacity Advisor commands

In addition to tasks that you can perform from the Capacity Advisor graphical user interface, Capacity Advisor provides several commands that can be run from the command line on an HP-UX orMicrosoft WindowsCMS. For specific information, see “Commandreference” (page 133).

22 Features

3 Key Capacity Advisor concepts

Capacity planning goals

Capacity planners are driven by multiple, sometimes conflicting, goals:

• Avoiding surprises

• Planning for the future

• Maintaining quality of service (see “Quality of service ” (page 23))

• Optimizing resource utilization

• Meeting internal and external security requirements

• Reducing costs

HP Insight Capacity Advisor software can help lessen the conflicts among goals by buttressing decisions with better information and more robust models. Effective use of Capacity Advisor can help you to double resource utilization in many situations.

The most important step in accomplishing capacity planning goals is to clearly understand the goals and the relationship among them. For example, optimizing response time may preclude trying to achieve the most efficient resource utilization, while optimizing resource utilization can compromise certain types of security. It is critical to understand the requirements as well as their priorities to provide best overall quality of service.

Quality of service

Typical aspects of quality of service include (but are not limited to):

• Availability

• Accessibility

• Integrity

• Performance

• Reliability

• Regulatory requirements

• Security

Quality of service is frequently defined in service level agreements between organizations.

Utilization monitor, calculator, and simulator

At a fundamental level, HP Insight Capacity Advisor software collects data from utilization monitoring daemons on systems and workloads. Using this data, Capacity Advisor can report on utilization of memory, core processing units, I/O bandwidth, and power. It can calculate what resource utilization would look like if the load were increased or decreased. Further, it can simulate what resource utilization might look like if loads are combined. This goes beyond simple addition of maximum loads to provide a dynamic addition of components based on the time sequence of utilization measurements taken from the actual loads running on real systems (see

“Peaks and sums” (page 25)).

Data handling for virtual machines

Data collected by Capacity Advisor for use in a profile viewer and historical utilization reports can vary from data reported by tools that run inside a virtual machine. These tools include HP-UX and Linux commands such as top, ps, and sar, and on Microsoft Windows, the task manager or perfmon.

This data can vary in a couple of ways. For example:

• Capacity Advisor usually reports CPU utilization that is greater than what is reported by

the VM guest. The operating system inside a virtual machine only knows about the CPU

Capacity planning goals 23

time used by threads that simulate the virtual processors. However, there are also threads that simulate the I/O cards and disks in the virtual system. Capacity Advisor collects data on all of these threads, which can cause the Capacity Advisor data for a specific virtual machine’s CPU utilization to be greater than the number of virtual CPU cores (vCPUs) associated with the virtual machine. Under certain load conditions, this can result in CPU utilization of more than 100% being reported for a virtual machine.

• Capacity Advisor data for VM host CPU utilization can be lower than what the guest

operating system reports, especially during periods of high utilization. Capacity Advisor records physical CPU utilization, which is the utilization of the actual cores on the VM host; that is, the total utilization reported by Capacity Advisor is based on the CPU time that the VM host allotted to each virtual machine. In contrast, the guest OS records virtual CPU utilization, which includes time when the VM wants to run, but is suspended while another VM is using the resources on the VM host.

Because Capacity Advisor corrects for these effects, the collected data has much less “noise” in it, and better reflects the CPU time that was actually used by any one VM.

Data normalization

Capacity Advisor makes certain assumptions in its data analysis. These assumptions include automatically adjusting for CPU platform differences based on a performance index. (The performance index is constructed by normalizing collected data to selected benchmarks for the source and destination systems).

If desired, you can add your own benchmark values to the database for use in calculating the performance index for workload moves among systems with the same or different architectures. (See capcustombenchmark in the Appendix B (page 133) for more information.

Memory, network IO, and disk IO usage are not normalized.

NOTE:

Using benchmarks for data normalization was introduced into the 6.0 release. This change affects workloads defined with static profiles in previous releases because they use clock speed to determine performance. Such workloads that are still used for planning should be redefined or edited in Capacity Advisor 6.x to improve the accuracy of the performance measure.

Resources for which Capacity Advisor collects data

HP Insight Capacity Advisor software collects and analyzes data for utilization of the following resources:

• CPU cores

• Memory

• Network I/O

• Disk I/O

• Power

Upper bounds

Upper bounds represent the maximum capacity of a resource for a given system. The maximum capacity is used in Capacity Advisor for items such as the network and disk I/O utilization graphs of systems, the graphs and values shown in a profile viewer, and the Capacity Advisor Smart Solver calculations.

CPU capacity The maximum CPU capacity of a system is the product of the number of CPU cores and the clock speed of the system.

Memory The maximum capacity of memory is the amount of memory on the system.

24 Key Capacity Advisor concepts

Network I/O The maximum capacity for network I/O defaults to the high-water mark (that is, the maximum observed or collected value) for network I/O. However, to set the upper bound to a value you choose, select Configure→Edit Network and Disk I/O Capacity... from the Visualization tab in Virtualization Manager.

Disk I/O The maximum capacity for disk I/O defaults to the high-water mark (that is, the maximum observed or collected value) for disk I/O. However, to set the upper bound to a value you choose, select Configure→Edit Network and Disk I/O Capacity... from the Visualization tab in Virtualization Manager.

Power Expected maximum and minimum power values for systems and workloads are computed using the CPU and power utilization data collected from HP Insight Control power management (IPM). However, these values can also be set manually from the Virtualization Manager menu bar (select Configure→Calibrate Power (All Selected Systems)...) for specified systems; or from the Capacity Advisor Edit Scenario: System tab menu bar (select Edit→Calibrate Power (All Selected Systems)...) for a specific scenario.

Measuring and analyzing resource utilization

In using Capacity Advisor, it is helpful to understand how the tool approaches sampling and data analysis, and the user-provided information that affects these.

Peaks and sums

Measuring utilization of computing resources is more complex than simply determining the maximum memory or processor utilization.

Sum of peaks An old standby in capacity planning is to simply take the peak of the two loads and use that to determine the maximum required capacity; this is the “sum of peaks”. While this will definitely provide a robust solution, it does not take into account the timing of the peak of the loads and may end up planning for more capacity than is actually used.

Peak of sums A more efficient planning solution, which is easily accomplishedusing HP Insight Capacity Advisor software, takes into account the timing of the maximum utilization peaks in the individual loads. By adding together utilization at each measured interval and then taking the maximum of the resulting time sequence, a more accurate measure of the required maximum resource can be determined. This can lead to cost savings when planning the resources required to consolidate loads onto new or existing servers.

Sampling interval

Where used, the Utilization Provider runs on each monitored system to collect information on resource utilization. At the CPU-clock cycle level, a processor is either busy or idle. For Capacity Advisor, the average utilization for each 5-minute (300 seconds) interval is stored. Therefore, peaks lasting less than 5 minutes are not visible.

Because each data point is the average of the five preceding minutes of values, this averaging tends to flatten the graphs, particularly when compared with real-time graphs in which each data point is the average of values from the 15 preceding seconds.

For data collected using an agentless solution, collection intervals can vary depending on values that you set and the number of machines in the collection.

Headroom

Headroom is the difference between the observed utilization on a system and the maximum available capacity. That is, the headroom of a system is the amount of additional capacity that can be used without violating the utilization limits of the applications running on that system. For example, if you have a system with 4 cores where you never want utilization to exceed 75%, and peak utilization is 1.75 cores, then headroom is 1.25 cores.

Measuring and analyzing resource utilization 25

Optimum headroom varies depending on size of system. While a single processor system might require 50% headroom to preserve reasonable response times, a 16-way system might have reasonable response times when loaded at 80%.

Adequate headroom can also depend heavily on the characteristics of the loads; highly interactive systems require much more headroom than those that can tolerate delays in response time; batch systems may get by with very little headroom at all.

Headroom star rating

Various reports and results show headroom star rating. Use the following chart to understand the headroom rating system.

MeaningStar image

Not only do all resources fit, but double the resource usage for any single workload could fit.

All resources fit, and at least 75% headroom for any single workload is available.

All resources fit, and at least 50% headroom for any single workload is available.

All resources fit, and at least 25% headroom for any single workload is available.

This star arrangement (1 green star) means that all resources fit in the system, but no or little headroom is available.

One or more resources do not fit in the system; the utilization limits are violated.

Data is not available for this system.

where

• resources can be CPU cores, memory, network I/O, and disk I/O. In the case of a virtual

machine, the number of CPU cores considered are those assigned to the VM, not the total number of cores on the VM host. The VM host clock speed, network capacity, and disk capacity are all inherited by the VM guest when it is moved onto the VM Host.

• fit means the utilization limits (see “Utilization limits ” (page 27)) are met

• headroom means “room for growth”

Interpreting the headroom star rating

Headroom star ratings for a host are a weighted average of all of the star ratings of the workloads on that host. The weighting tends to give the highest weight to the lowest rating. One low rating can dramatically lower the rating for the entire host.

In the case of a VM host, the star ratings account for how well the workloads fit into their virtual machines, as well as how well the virtual machines fit on the VM host. The rating for the VM host will be low if any of the virtual machines are too small for their workloads.

Interpreting the star rating given by the HP Smart Solver

When using the Smart Solver to find a plan to convert physical systems to virtual machines, consider the following factors that can adversely affect the Smart Solver results.

• The addition of a virtualization overhead multiplier to a VM will often reduce the number

of stars for that workload by 1 or 2 stars.

• The clock speed of the VM host may be slower than the original physical system. Work that

was done by 1 core at 2.6 GHz, may require 2 cores when placed on 2.1 GHz VM host.

26 Key Capacity Advisor concepts

You can avoid having the Smart Solver produce inaccurate or useless results by re-sizing your systems before running the Smart Solver. If either of the above conditions exist in your situation, consider increasing the number of cores on your simulated physical systems before running the Smart Solver. (Select What-if Actions→Edit System... on the System tab on the Edit Scenario screen.) If you change the number of cores from 1 to 2 before consolidating, for example, the resulting virtual machines will have enough cores to cover the virtualization overhead or a slower VM host.

Re-sizing the virtual machines after running the Smart Solver can be less effort, as you only have to re-size the VMs that have fewer stars than your desired goal. After adding more cores to the VMs for which CPU resources are too tight, you can rerun Smart Solver to balance the load on the VM hosts to improve the solution a bit more.

TIP: Use a Scenario comparison report to compare the headroom starsrating for saved scenarios.

Missing or invalid data

Data collected by Capacity Advisor is used in the scenarios you create and manipulate. During an interval when no data was collected, the data is considered missing (data may not have been collected, for example, because a system was down during data collection). Invalidated (or invalid) data is data that you have marked as invalid.

For each metric about a system or workload, if a significant amount of data is missing or invalid, the metric is followed by asterisks with the following meaning:

• [blank] : 91% to 100% of data is valid.

• * : 51% to 90% of data is valid.

• ** : 11% to 50% of data is valid.

• *** : Less than 10% of data is valid.

• N/A: There is no valid data.

Thus, metrics asterisks are considered useful and reliable for analysis.

NOTE:

In some situations, where time or time zones on a server are incorrect, it may appear that only old data is available in the data collection. For more information on this topic, see the section “Handling Old Data” in Appendix H (page 175).

The affect of invalid data on HP Smart Solver solution finding

The HP Smart Solver is set to ignore workloads having >25% invalid data; that is, the Solver will not consider these workloads within an automated solution. This percentage can be adjusted, with a corresponding affect on the Solver's ability to include workloads in its solution calculations. For more information on what to do with workloads that violate this limit, see “When the

threshold for invalid data is exceeded” (page 181).

Utilization limits

The default utilization limits used globally across Capacity Advisorin the absence of user-defined limits are the following:

• CPU utilization cannot exceed 70% of the capacity for more than 15 minutes at a time.

(Seventy percent is used as adefault for CPU utilization as it providesacceptable performance with a minimum of queuing in jobs.)

• Memory utilization cannot exceed 100% of the capacity. Typically memory should be set at

a value <100% to allow for memory use by the dynamic buffer cache and operating system activity.

Measuring and analyzing resource utilization 27

(For more information on how utilization is calculated for each resource, see Appendix C (page 153) .)

Specifying utilization limits

There are three building blocks to specifying a utilization limit:

• The limit The maximum percentage or absolute amount of a resource allowed to be used

by a workload. For example, a CPU utilization limit might be “not above 90%” utilization.

• The resource Utilization limits are applied to specific resources:

— CPU cores — memory — network I/O bandwidth — disk I/O bandwidth

TIP:

You can specify more than one utilization limit for a resource Using the Utilization Limits

Editor, you can add multiple settings for a resource. For example, you can create multiple different utilization limits for CPU cores by varying percentage and allowed duration for each limit. Multiple limits for CPU cores could look like this:

— Utilization can exceed 90 percent of assigned cores 0 percent of the time — Utilization can exceed 85 percent of assigned cores for a maximum of 5 minutes duration

Not specifying a limit allows HP Smart Solver to over-provision systems To achieve best results with the Smart Solver, it is better to set specific limits, rather than to depend on the default settings for limits to provide the best fit.

• The time criteria You can specify the time portion of a utilization limit in either of two

ways: — Sustained (consecutive) time limits — Percentage of time limits

Sustained time limits

A sustained limit specifies a limit where the resource cannot exceed that utilization limit for X consecutive minutes. For example, if X is 20, this means that the resource cannot exceed the

utilization limit for 20 consecutive minutes.

Because the Capacity Advisor collects data samples every 5 minutes, the time X for the sustained limit must be a multiple of 5 minutes; the minimum for X is 0 minutes.

Percentage of time limits

A percentage of time limit specifies that the resource cannot exceed the limit for more than the designated percent of time, where percent of time is related to the percentile utilization ranges in the Capacity Advisor data.

Given that there are about 10,000 minutes in a week, 3% of the time is roughly 300 minutes (3% of 10,000). These 300 minutes total to 5 hours per week. Below is a table relating percentages of time to hours per week, which may help you in specifying percent of time utilization limits.

Table 3-1 Percent of time conversions

28 Key Capacity Advisor concepts

Hours/ WeekMinutes/ WeekPercent of Time

Hours/Day

(24–hour day)

.241.68100.81

.483.36201.62

Table 3-1 Percent of time conversions (continued)

Understanding utilization limit messages

Percentage of allocation

The utilization limit messages are shown as a percentage of allocation, where allocation is subset of the given hardware for the system the workload is running on. For example, for a 1-core system, the allocation is 1 CPU. The CPU utilization limit of 50% would mean 50% of 1 core, or .5 cores. However, this percentage changes when the hardware (allocation) changes. If 2 additional cores are added (say through dynamic CPU migration with vPars), the CPU utilization limit of 50% would mean 50% of 3 cores, or 1.5 cores.

Allocation values for network IO and disk IO are derived from either the highest observed value for that metric over the entire range of data collected, or a user-specified maximum value.

The allocation values fornetwork anddisk may be updated each timeutilization data is collected from the system using the capcollect command. If a new high observed value occurs during the time period collected, the network or disk allocation value for the system is increased to reflect it. This increased value then affects any network or disk utilization limits for workloads on that system. The current allocation values for a system are displayed on the Profile Viewer page under Platform Characteristics.

Hours/ WeekMinutes/ WeekPercent of Time

Hours/Day

(24–hour day)

.725.04302.43

1.208.40504.05

2.4016.81008.010

3.6025.21512.015

4.8033.62016.020

6.0042.02520.025

7.2050.43,024.030

24.00168.0010080.0100

With sustained limits

A sustained utilization limit could be set such that CPU utilization cannot exceed 50% of allocation for 20 consecutive minutes, where the allocation of hardware is based upon a 3–core system. The utilization limit message would read:

CPU utilization may not exceed 50% of allocation or 1.5 cores for more than 20 minutes.

With percentage of time limits

A percentage of time utilization limit could be set such that CPU utilization cannot exceed 50% of allocation for more 10% of the time, where the allocation is based upon a 3-core system. The utilization limit message would read:

CPU utilization may not exceed 50% of allocation or 1.5 cores for more than 10% of the time.

Scope of utilization limits

Utilization limits can be set to apply broadly or narrowly within the Capacity Advisor user interface:

Measuring and analyzing resource utilization 29

• Globally. These limits apply to every workload, wherever workloads are analyzed.

• By Workload. These limits apply to one specific workload, wherever that workload is analyzed.

• Scenario-wide. These limits apply to every workload within one specific scenario.

• By Scenario Workload. These limits apply to one specific workload within one specific scenario.

When a workload falls within more than one scope, only the more specific one applies, as shown in the table below.

You can disable a more specific scope where you do not want a specific scope to apply.

Table 3-2 Scope of utilization limits

OverridesDescriptionLimitsScope

global

More Local

Global Utilization Limit

Workload Utilization Limit

Scenario Utilization Limit

Scenario Workload Utilization Limit

Adjusting power

With the cost of power increasing, power has become a resource that you may want to measure and manage. Within HP Insight Dynamics suite, power metrics, graphs and reports are displayed for actual systems and for systems within capacity planning scenarios.

NOTE: You can calibrate power for physical systems (including virtual partitions), but not for

virtual machines.

utilization limit is not provided.

• Cannot be disabled

utilization limit is provided.

• Can be enabled or disabled

• Applies to all workloads within a scenario for which a more

specific utilization limit is not provided.

• Can be enabled or disabled

• Applies to a specific workload within a scenario.

• Can be enabled or disabled

• Nothing• Applies to all workloads for which a more specific

• Global• Applies to a specific workload unless a more specific

• Global

• Workload

• Global

• Workload

• Scenario

To make the power data as accurate as possible, you are provided with the opportunity to calibrate power usage using data collected from actual systems, or you can supply your own values.

Power cap data

The power cap function on some HP hardware allows you to set a limit in watts for a server, and the server will ensure that its power consumption never exceeds that limit. That, in turn, can reduce power and cooling costs. It can also help to ensure that a data center does not draw more energy from the power grid than it is allocated to use, reducing the likelihood of experiencing unplanned downtime necessary to stay within a power allocation limit.

The potential drawback to not setting the power cap high enough is degraded hardware response time, which an administrator can rectify by adjusting the cap -- or accept in exchange for the associated savings. For detailed information on how power caps work in HP hardware, see HP

Insight Power Manager.

Capacity Advisor gives you insight into the current power cap for HP systems that have these options available and operating. Within Capacity Advisor, you can generate a power report to see historic power usage data that includes power cap settings and estimated enforcement time.

The following image shows an example of the power cap data included in a Capacity Advisor power report.

30 Key Capacity Advisor concepts

+ 168 hidden pages