Provision elastic cloud load generators

This section describes how to reserve (or provision) a load generator on the cloud with a specific configuration for running tests.

Alternatively you can specify an image that is included in a custom configuration. For details, see Create custom image for cloud hosts.

Configure a test with cloud-based load generators

This task describes how to configure a performance test with elastic cloud load generators. You can perform this from the LoadRunner Enterprise user interface or the REST API.

Tip: See the LoadRunner Enterprise video gallery for a video on how to run tests on load generators that are dynamically provisioned on AWS.

Configure a test from the LoadRunner Enterprise user interface

  1. Prerequisites

    • You must have at least one cloud account and at least one host template linked to your project. For details, see Cloud accounts and Host templates.

    • Make sure the Cloud Network Settings contain the required port and certificate settings. For details, see Configure cloud network settings.

    • When provisioning hosts for Amazon EC2 cloud accounts, you need to supply the Security Group and Key Pair. For details, see Security group rules.

      Note: Security groups and key pairs are created separately for each cloud location. Make sure that you have configured the necessary security groups and key pairs for the locations you provision from.

    • Other considerations when provisioning hosts:

      Location The location where you provision hosts can impact the effectiveness of your testing. Make sure to select a cloud location that matches your testing needs.
      Host templates are created separately for each location; therefore a template that exists in one location may not exist in a different location. Make sure that you know which locations contain the templates you need or that you add your desired templates to the locations you intend to provision from.
      Remote Connectivity

      If you need to connect remotely to the provisioned hosts, make sure to select a key pair for which you have the private key and to select a security group that allows remote connections.

  2. Provision load generators in a specific subnet in a VPC

    LoadRunner Enterprise supports provisioning load generators in Amazon default and custom virtual private cloud (VPC).

    If you are using a specific subnet in a VPC, you need to supply the subnet in a VPC in a host template. For details on how to set this up, see Provision load generators in a specific subnet in a VPC.

  3. Assign elastic cloud load generators to the groups for the test

    1.  In the top banner, click the module name or the dropdown arrow and select Test Management > Tests & Scripts. Select a performance test in the test management tree, and click Edit Test.

    2. In the Performance Test Designer window, click Groups and Workload. On the Groups pane toolbar, click the Assign LGs button to open the Assign Groups to Load Generators dialog box. For details, see Distribute load generators among Vuser groups.

    3. In the Groups column, select the groups to use in the test.

    4. In the Load Generators column, expand the Elastic Cloud pane.

    5. Click the Add button to open the Add Elastic Cloud LG dialog box, which enables you to add more cloud load generators for the test.

    6. For each host template you select, enter the number of load generators you want to add to the test. You can filter the list of host templates in the grid by account, location, size, and/or operating system.

      For more details, see Assign elastic hosts to a test.

    7. Click Add to add the selected load generators and close the dialog box, or Add & Another to add and keep the dialog box open enabling you to add additional cloud load generators.

    8. After adding the cloud load generators you require, select the load generators that you want to assign to the selected groups.

  4. Click the module name or the dropdown arrow in the top banner, and select More Tools > Controller Options. In the Timeout tab, enter 200 in the Connect operation box. Click Save.

  5. If your LoadRunner Enterprise administrator did not set a post-run action, set it to collate results.

    1. Click Run and in the Run Test dialog box click the New Timeslot tab.

    2. Select Collate results or Collate and analyze to ensure that run results are not lost after a performance test ends.

      You can also configure LoadRunner Enterprise to collate results during runtime in the general test options. For details, see Configure collate options.

      For details, see Run a test.

    Note: If a test is run without setting a collate results option, the results are lost because all elastic cloud load generators are de-provisioned immediately after the test run finishes.

  6. Run the performance test

    Save and then run the test to check its performance. For details on running a test, see Run performance tests.

    When the test has finished, the elastic cloud hosts are automatically de-provisioned.

Configure a test using the LoadRunner Enterprise REST API

  1. Create or update performance tests with elastic cloud-based hosts, by selecting dynamic type hosts for your groups.

    Add elastic load generator hosts named CLOUD1, CLOUD2, CLOUD<n+1> to the <Host> XML field, equivalent to the number of cloud hosts that you want to provision.

    Copy code

    Example:

    <Hosts>   
        <Host>      
           <Name>CLOUD1</Name>     
           <Type>cloud</Type>
           <HostTemplateId>11</HostTemplateId>
        </Host>
        <Host>
           <Name>CLOUD2</Name>
           <Type>cloud</Type>
           <HostTemplateId>912</HostTemplateId>
        </Host>
    </Hosts>

    For API details, see test entity XML in the LoadRunner EnterpriseAdministration REST API Guide.

  2. Run the performance test

    1. Prerequisites

      Make sure that you have created a performance test, and configured it to use non-specific (automatch) load generators. You can also use specific load generators, but then you need to update your test to use those load generators (this can be done using tests/{ID} in the LoadRunner Enterprise REST API Guide).

      You can create a test either from the LoadRunner Enterprise Administration REST API (see tests in the LoadRunner Enterprise REST API Guide), or manually from the LoadRunner Enterprise user interface (see Design a test).

    2. Run the performance test from the LoadRunner Enterprise REST API (see Start a Test Run in the LoadRunner Enterprise REST API Guide).

  3. Gather results

    When the performance test has finished running, collate (and analyze) performance test data using the REST API. For API details, see collate and analyze in the LoadRunner Enterprise REST API Guide.

Back to top

Provision load generators in a specific subnet in a VPC

If you are using a specific subnet in a VPC, you need to supply the subnet in a VPC in a host template.

  1. Create a VPC from Amazon Web Services.

    1. In the subnet configuration, set Auto-assign Public IP as required.

      Note: If Auto-assign Public IP is set to Disable, the load generators have a private IP only, and need to establish a trust between the corporate network and the Amazon Web Services VPC.

    2. Create a relevant security group for the VPC.

  2. Create a host template.

    Create a host template and set the subnet ID in the VPC Subnet Id field of the template.

    For details, see Create an elastic cloud host template.

  3. Provision a load generator from LoadRunner Enterprise using one of the templates created for the VPC using the required subnet ID.

  4. Select a security group that was created for the selected VPC.

Back to top

Notes and limitations

The following notes and limitations apply to elastic cloud load generators:

Permissions

A user must have Provision Cloud host permissions to be able to run a test with Elastic Cloud load generators. If you do not have these permissions, either contact your system administrator, or assign another type of load generator for the test.

Timeslot

Since provisioning Elastic Cloud load generators can be time-consuming, we recommend setting a timeslot duration of at least one hour.
Host templates

Before deleting host templates, make sure there are no assigned projects with tests using the templates (deleting a host template that is used by a test causes the test to fail).

Result collation For issues when running tests without collating results, or if collation fails, see Retain run results after a test ends above.
Network Virtualization

  • The NV port must be opened when using elastic Cloud hosts.

  • NV graphs might not be displayed online during a test run, but should be available offline.

  • While latency applied by NV emulation does not impact WAN latency, latency for an application under test can be significantly higher than specified in the virtual location.
Firewall

Running over a firewall is not supported.
Amazon EC2

When adding an Amazon EC2 cloud account to LoadRunner Enterprise, the available images are retrieved from Amazon EC2. If any of the Amazon EC2 images in the templates subsequently become invalid (for example, if an image is updated), and you try to provision Load Generators using a cloud template that points to an invalid Amazon EC2 image ID, the operation fails.

Resolution: Identify the template that points to an invalid Amazon EC2 image, and create a custom template with a valid Amazon EC2 image.
Multiple installations

If you have multiple LoadRunner Enterprise installations using the same project, the cloud load generator cleanup task removes all load generators regardless of which LoadRunner Enterprise application provisioned them. This can happen during migration when both the original and migrated projects coexist for a period of time.

Resolution: To avoid cleaning load generators that were provisioned by another LoadRunner Enterprise application, perform the following:

  1. Go to the <LRE_server_installdir>\LRE_BACKEND\appsettings.defaults.json file.

  2. Add the LRE_INSTALLATION tag to the "InstanceTags" section under the relevant cloud provider.

  3. Change the tag value on every environment that you want to prevent the cleanup job.

Copy code

Example of tag added to an AWS cloud provider:

"AWS":{
"InstanceTags": [
{
"Key": "LRE_INSTALLATION",
"Value": "LRE_DEV"
},
{
"Key": "CUSTOMTAG",
"Value": "CUSTOMTAG_VALUE"
}
],
"ElasticIpsTag": {
"Key": "LRE_TEST_TAG",
"Value": "LRE_TEST_TAG_VALUE"
},
"Endpoints": {
"Standard": 
"us-west-1","USGov": "us-gov-west-1"
}
},
Maximum number of load generators per template

By default, the maximum number of provisioned load generators for a single host template assignment is 199.

To change the default, assign a value to the MaxHostLimitOneTemplate variable in the Cloud section of the following file: <Server_Installdir>\LRE_BACKEND\appsettings.defaults.json. For example:

Copy code 
"Cloud":{
 "MaxHostLimitOneTemplate": 301,

Back to top

See also: