ONTAP Select Performance Characterization Tool

ONTAP Select Performance Characterization
Tool
Configuration Guide
Aditya Siddavatam, NetApp
September 2017
Abstract
The NetApp® ONTAP® Select Performance Characterization Tool is a packaged solution that
allows the user to initiate I/O workloads and perform metrics collection, archiving, and
graphing in the same virtual appliance. This document was written for NetApp engineers who
need a faster and more efficient performance evaluation tool for developing NetApp ONTAP
Select systems.
TABLE OF CONTENTS
1 Introduction ..................................................................................................................................... 3
2 Deploying the ONTAP Select Performance Characterization Tool ................................................ 3
3 Configuring Network, Tools, and I/O Components ........................................................................ 3

3.1 Configuring the Network............................................................................................................................ 4
3.2 Configuring Tools ..................................................................................................................................... 7
4 Initiating I/O Workloads ................................................................................................................. 19

4.1 Validating All Configurations.................................................................................................................... 19
4.2 Starting an I/O Workload ......................................................................................................................... 20
4.3 Monitoring Workload Clients.................................................................................................................... 20
4.4 Collecting Performance Data................................................................................................................... 21
5 Exiting and Relaunching the Tool................................................................................................. 22
2 ONTAP Select Performance Characterization Tool © 2017 NetApp, Inc. All rights reserved.
1 Introduction
The NetApp ONTAP Select Performance Characterization Tool (PCT) is a packaged solution that allows
the user to initiate I/O workloads and perform metrics collection, archiving, and graphing in the same
virtual appliance.
This tool is designed to be easy to deploy and operate through the command-line interface. It is based on
the following products:
 FIO. I/O workload generator.
 Harvest. Dashboards and data collection.
 Graphite. Metrics database.
 Grafana. Web-UI for dashboards.
This document was written for NetApp engineers who need a faster and more efficient performance
evaluation tool for developing NetApp ONTAP Select systems.
2 Deploying the ONTAP Select Performance Characterization Tool

The NetApp ONTAP Select PCT virtual appliance can be installed with the vSphere web client by using
the Open Virtual Machine Format (OVF). The deployment procedure is the same as for a standard OVF
installation.
1. Deploy the OVF template from a local file system accessible to the vSphere Client machine or from
the web URL.
2. Boot the appliance after the deployment is finished.
3. After the appliance boots, you can assign a static or Dynamic Host Configuration Protocol (DHCP) IP
address to the virtual machine (VM). If a DHCP server is available on the network, the appliance can
obtain IP configuration details automatically. Otherwise, you must configure the IP information in the
console. Th configuration of IP details is described in the section “Configuring Network, Tools, and
I/O Components.”
4. The default login credentials for the VM are as follows:
 Username: root
 Password: root
3 Configuring Network, Tools, and I/O Components

After you install the PCT, you must provide configuration details for network setup, the I/O components,
and the tools needed to perform I/O workloads and data analysis. The first time that you log in to the VM,
guidelines for creating I/O components are printed out. In addition, the main menu presents the following
options:
Select option 1 from the main menu to configure the following components before initiating an I/O
workload.
 Network. Configure an IP address to communicate with external network components such as an
ONTAP controller, the load clients, and so on.
 Tools. Configure tools to perform data collection and analysis.
 I/O components. Create and set up I/O components.
3.1 Configuring the Network

Configure either a static IP address or a DHCP IP address for the VM. After the address is configured,
you can communicate with the outside network, such as an ONTAP controller, load clients, and so on, to
run commands or perform I/O operations.
To configure the network, select option 1 from the Sub-Menu for Configuring Network, Tools, and I/O
Components. The primary interface for both management and data networks. Please configure eth1 if a
separate network is needed for data network.
Then enter the following information:
 IP Address type. Enter 1 for a static IP address or 2 for a DHCP IP address.
 IP Address. Enter an IP address for the VM hosting the PCT.
 Netmask. Enter netmask for the IP.
 Gateway. Enter gateway for the IP.
 Domain. Enter the DNS domain name.
 DNS1. Enter the domain name server (DNS) IP.
 DNS2 (if needed). Enter a secondary domain name server(DNS) IP.
You can optionally configure a second network interface for the appliance. This is useful if the Select
cluster’s data network is separate from the network used to manage the PCT appliance
 IP Address. Enter an IP address for the VM hosting the PCT.
 Netmask. Enter netmask for the IP.
 Gateway. Enter gateway for the IP.
 Domain. Enter the DNS domain name.
 DNS1. Enter the domain name server (DNS) IP.
 DNS2 (if needed). Enter a secondary domain name server(DNS) IP.
Note: Use a host name or IP address to make sure that the domain name and DNS that you enter can
communicate with the ONTAP controller and the client(s) used to initiate I/O workload.
3.2 Configuring Tools

After you configure the network, you must configure the tools (Harvest, Graphite, and Grafana) needed to
perform data collection, archiving, and graphing. To complete this step, you must have the available
hostname or IP address of the ONTAP cluster against which the performance tests will run.
Configuring a Harvest User

Harvest requires login credentials for the ONTAP controllers. Although a generic admin account can be
used, it is a best practice to create dedicated monitoring accounts with the correct privilege levels.
The following sections contain the steps necessary to configure access for a dedicated user.
Configuring a Role
To configure a role, run the following commands:
security login role create -role netapp-harvest-role -access readonly -cmddirname "version"
security login role create -role netapp-harvest-role -access readonly -cmddirname "cluster
identity show"
security login role create -role netapp-harvest-role -access readonly -cmddirname "cluster show"
security login role create -role netapp-harvest-role -access readonly -cmddirname "system node
show"
security login role create -role netapp-harvest-role -access readonly -cmddirname "statistics"
security login role create -role netapp-harvest-role -access readonly -cmddirname "lun show"
security login role create -role netapp-harvest-role -access readonly -cmddirname "network
interface show"
security login role create -role netapp-harvest-role -access readonly -cmddirname "qos workload
show"
Configuring a User
 To configure a user for Data ONTAP, run the following commands:

security login create -user-or-group-name netapp-harvest -application ontapi -role
netapp-harvest-role -authmethod password
Configuring and Starting the Harvest, Graphite, and Grafana Processes

To configure Harvest, Graphite, and Grafana, select option 2 from the Sub-Menu for Configuring Network,
Tools, and I/O Components and enter the following information:
 Node/cluster management host name or IP address. Enter the ONTAP controller IP or host name.
 Username. Enter the username created for Harvest.
 Password. Enter the password.
 Harvest group name. Enter a group name. The user can choose any name
Configuring the Time Zone
As a part of configuration, you must configure the time zone for the VM so that the performance graphs
are synchronized with the I/O workloads and the ONTAP cluster under test.
There is no option for configuring the time zone in Grafana. Therefore, graphs are listed based on the
browser time zone rather than on the workload client’s time zone. To make sure that graphs are in sync
with the workload client(s), NetApp recommends that you configure the system time zone from which you
are viewing Grafana to match the workload client time zone.
After you enter the group name, the Configuring Timezone screen appears. To configure the time zone,
complete the following steps:
1. Select the appropriate continent and country.
2. Enter 1 if the resulting information is correct and 2 if it is not.
Configuring the Data and Time
After setting the time zone, you can set the date and time.
1. Enter 1 if you would like to set the date and time.
2. Enter the date and time in the following formats: YYYY-MM-DD and HH:MM:SS.
When the time zone, the date, and the time have been set, the Graphite, Graphana, and Harvest
processes start.
Verifying Harvest Status

To check the status of Harvest process after configuration completion please follow the below steps:
1. The Harvest log path can be found at the following location. The log name is generated as
<ONTAP_Cluster_Name>_netapp-harvest.log.
/opt/netapp-harvest/log/
2. To find the Harvest process ID, run the following command:

pgrep netapp-worker
3. To find the Harvest status, run one of the following commands:

/opt/netapp-harvest/netapp-manager -status
service netapp-harvest status
The output from these commands tells you if Harvest is collecting data. If the status is in the RUNNING
state, then Harvest is collecting data. If NOT RUNNING or STOPPED appears, then there might be some
issue that is preventing Harvest from collecting data. For more information on the failure, refer to the logs
in the log path from step 1.
Configuring I/O Components
To create new components, such as a volume, a storage virtual machine (SVM, also referred to as a
vserver within the ONTAP CLI), or a route, you can leverage existing ONTAP components. You can
then update the component information in the I/O workload configuration file used by the I/O workload
tool.
As mentioned previously, you can use the ONTAP cluster management information from the Harvest
configuration to communicate with ONTAP and create components on top of ONTAP.
As a first step, see the guidelines for creating I/O components. These are printed out the first time that
you log in to the VM and they are also presented in the CLI, as can be seen in the following screenshot. If
you would like to reprint the guidelines, select option 3 from the main menu.
This example uses the client that initiates the I/O workload as the first load client.
Also, it is recommended to have 4 load clients, which includes the client that will be used for starting the
I/O workload and it will be considered as first load client. Other 3 load clients are preferably deployed
with PCT ova. If only one load client is provided then workload will be distributed between two load
clients(first load client-PCT, second load client-provided client)
Creating New Components

As a prerequisite, you should have at least one data aggregate with sufficient free space. To create new
components, complete the following steps:
1. Select option 3 from the Sub-Menu for Configuring Network, Tools, and I/O Components.
2. Select option 1 under Configuring I/O Components.
3. Enter the ONTAP cluster credentials, and then enter the name of the aggregate with the required free
space to create the I/O components. After you provide this information, the SVMs and volumes are
created as shown below:
4. After the volumes and SVMs are created, the wizard proceeds with data LIF creation. Provide the
data IP, data port name, netmask, and gateway for the data LIF. After you enter this information, the
network interface, the network route, the NFS service, and the export policy rule are created.
5. Depending on the number of volumes, the wizard asks you to provide additional load client details for
starting I/O. Once complete, a ping test will be done to each client.
Cleanup existing components
To clean up any I/O components previously created using the tool:
Select option 1: Configure Network, Tools and IO components.
Select option 3: Configure IO components
Select option 3: Clean up existing IO components created by the tool.
Only IO components previously created with the tool can be cleaned up using this option
Using Existing Components
If you have already created I/O components, the following steps allow you to use them in the PCT:
1. Select option 3 from the Sub-Menu for Configuring Network, Tools, and I/O Components and then
select option 2 to use existing I/O components:
2. After selecting option 2, you are prompted to provide additional load client details, depending upon
the number of volumes.
4 Initiating I/O Workloads
After performing the configuration of the network, tools, and I/O components, you can start the I/O
workload.
Select option 2 from the Main Menu to access the Sub-Menu for Initiating I/O Workload.
4.1 Validating All Configurations

Prior to running an I/O workload, you should first verify whether all configuration files have been
successfully created.
To do so, select option 1 from the Sub-Menu for Initializing I/O Workload.
4.2 Starting an I/O Workload
After verifying that all configuration files have been created, you can then trigger an I/O workload using
the FIO tool. The configuration used in this document performs the following processes:
 64K sequential read
 4K random read
 64K sequential write
 4K 50% random read and 50% random write
 4K random write
Select option 2 from the Sub-Menu for Initializing I/O Workload to run the I/O workload. Afterwards, the
Graphite and Grafana URL information and credentials are presented. You can then log in and view the
resulting graphs.
The Graphite URL and login credentials are as follows:
 URL: http://<IP_Address>:81
 Username: admin
 Password: admin
The Grafana URL and login credentials are as follows:
 Username: admin
 Password: admin
 URL: https://<IP_Address>
Note: The <IP_Address> is the VM IP address on which you are performing these configurations.
Note: NetApp recommends using the Chrome browser to view Graphite and Grafana.
4.3 Monitoring Workload Clients

To monitor workload clients, access the following logs in PCT VM:
 The FIO test log path can be found at the following location, and the log name is generated as
io_workload.out:
/opt/netapp-harvest/log/io_workload.out
This log contains the steps that were performed on the clients to run the workload.
 The client workload log path can be found at the following location and the log directory name is
generated as <DATA_VSERVER>-<TIMESTAMP>:
/root/fio_test_results/
This log describes the workload that is being executed and its corresponding I/O statistics.
 The I/O components log path can be found at the following location and the log name is generated as
<TIMESTAMP>.log:
/opt/netapp-harvest/log/io_components_logs
This log contains the debug information regarding I/O components that have been created on the
config.
This tool does not enable or disable compression and deduplication. It is up to the user to enable or
disable these features.
As mentioned in the guidelines, the PCT VM is used as one of the I/O clients to run the workload.
Therefore, make sure that the network has been configured properly and that the workload clients are
able to communicate with the ONTAP controller, because mount operations are performed. Make sure
that the workload clients have the FIO tool under the following path:
/usr/software/test/bin/
This configuration is necessary if you use any external clients.

In the future, you might want to make changes to the network, tools, or I/O workload configuration file(s).
For example, you might want to update the DNS, the domain name, the cluster management IP address
or host name, client details, or the data SVM. To do so, return to the appropriate sub-menu and select the
relevant option. Doing so overwrites the existing data.
4.4 Collecting Performance Data

You can use Perfstat or an equivalent performance collection tool on an ONTAP Select cluster when the
PCT VM starts running a maximum peak workload.
The tool calculates and displays peak load details, the average throughput, and the latency after
completing each I/O workload. In the example given below, the tool has finished running a sequential-
read I/O workload and is displaying the details for each volume.
Afterwards, the tool initiates the same workload(above example sequential read)with IO attributes
that contributed peak performance(avg throughput) on all the volumes for approximately 10 minutes. You
should start Perfstat or an equivalent performance collection tool on the ONTAP Select cluster at this time.
In a similar manner, you can collect performance data for each I/O workload(random-read,
sequential-read, random-read-write, random-write).
5 Exiting and Relaunching the Tool

Select option 4 from main menu to exit the tool.
To relaunch the main menu after exiting, run the following command:
/opt/netapp-harvest/startup.sh
Refer to the Interoperability Matrix Tool (IMT) on the NetApp Support site to validate that the exact
product and feature versions described in this document are supported for your specific environment. The
NetApp IMT defines the product components and versions that can be used to construct configurations
that are supported by NetApp. Specific results depend on each customer’s installation in accordance with
published specifications.
Copyright Information
Copyright © 1994–2017 NetApp, Inc. All rights reserved. Printed in the U.S. No part of this document
covered by copyright may be reproduced in any form or by any means—graphic, electronic, or
mechanical, including photocopying, recording, taping, or storage in an electronic retrieval system—
without prior written permission of the copyright owner.
Software derived from copyrighted NetApp material is subject to the following license and disclaimer:
THIS SOFTWARE IS PROVIDED BY NETAPP “AS IS” AND WITHOUT ANY EXPRESS OR IMPLIED
WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF
MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE, WHICH ARE HEREBY
DISCLAIMED. IN NO EVENT SHALL NETAPP BE LIABLE FOR ANY DIRECT, INDIRECT,
INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR
PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF
LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR
OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF
THE POSSIBILITY OF SUCH DAMAGE.
NetApp reserves the right to change any products described herein at any time, and without notice.
NetApp assumes no responsibility or liability arising from the use of products described herein, except as
expressly agreed to in writing by NetApp. The use or purchase of this product does not convey a license
under any patent rights, trademark rights, or any other intellectual property rights of NetApp.
The product described in this manual may be protected by one or more U.S. patents, foreign patents, or
pending applications.
RESTRICTED RIGHTS LEGEND: Use, duplication, or disclosure by the government is subject to
restrictions as set forth in subparagraph (c)(1)(ii) of the Rights in Technical Data and Computer Software
clause at DFARS 252.277-7103 (October 1988) and FAR 52-227-19 (June 1987).
Trademark Information
NETAPP, the NETAPP logo, and the marks listed at http://www.netapp.com/TM are trademarks of
NetApp, Inc. Other company and product names may be trademarks of their respective owners.

ONTAP Select Performance Characterization Tool

Caricato da

Informazioni sul documento

Copyright

Formati disponibili

Condividi questo documento

Condividi o incorpora il documento

Opzioni di condivisione

Hai trovato utile questo documento?

Questo contenuto è inappropriato?

Copyright:

Formati disponibili

ONTAP Select Performance Characterization Tool

Caricato da

Copyright:

Formati disponibili

ONTAP Select Performance Characterization

2 Deploying the ONTAP Select Performance Characterization Tool ................................................ 3

3 Configuring Network, Tools, and I/O Components ........................................................................ 3

3.2 Configuring Tools ..................................................................................................................................... 7

4 Initiating I/O Workloads ................................................................................................................. 19

4.4 Collecting Performance Data................................................................................................................... 21

5 Exiting and Relaunching the Tool................................................................................................. 22

2 Deploying the ONTAP Select Performance Characterization Tool

3 Configuring Network, Tools, and I/O Components

3.1 Configuring the Network

3.2 Configuring Tools

Configuring a Harvest User

 To configure a user for Data ONTAP, run the following commands:

Configuring and Starting the Harvest, Graphite, and Grafana Processes

Verifying Harvest Status

2. To find the Harvest process ID, run the following command:

3. To find the Harvest status, run one of the following commands:

Creating New Components

4.1 Validating All Configurations

4.3 Monitoring Workload Clients

This configuration is necessary if you use any external clients.

4.4 Collecting Performance Data

5 Exiting and Relaunching the Tool

Potrebbero piacerti anche