Main Content

Configure for HPC Pack

Configure Cluster for Microsoft HPC Pack

Follow these instruction to configure your MATLAB® Parallel Server™ installation to work with Microsoft® HPC Pack or Compute Cluster Server (CCS). In the following instructions, matlabroot refers to the MATLAB installation location.

Supported versions: MATLAB Compute Cluster Server 2003, Windows® HPC Server 2008, Windows HPC Server 2008 R2, Microsoft HPC Pack 2012, Microsoft HPC Pack 2012 R2, Microsoft HPC Pack 2016, and Microsoft HPC Pack 2019.

Note

If you are using an HPC Pack in a network share installation, the network share location must be in the “Intranet” zone. You might need to adjust the Internet Options for your cluster nodes and add the network share location to the list of Intranet sites.

  1. Log in on the cluster head node as a user with administrator privileges.

  2. Open a command window with administrator privileges and run the following file command

    matlabroot\toolbox\parallel\bin\MicrosoftHPCServerSetup.bat -cluster 

    If you are using an HPC Pack in a network share installation, this command performs some of the setup required for all machines in the cluster. The location of the MATLAB installation must be the same on every cluster node. If you are not using an HPC Pack in a network share installation, then you must run this command on every cluster node.

    Note

    If you need to override the script default values, modify the values defined in MicrosoftHPCServerSetup.xml before running MicrosoftHPCServerSetup.bat. Use the -def_file argument to the script when using a MicrosoftHPCServerSetup.xml file in a custom location. For example:

    MicrosoftHPCServerSetup.bat -cluster -def_file <filename>

    You modify the file only on the node where you actually run the script.

    An example of one of the values you might set is for CLUSTER_NAME. If you provide a friendly name for the cluster in this parameter, it is recognized by MATLAB’s discover clusters feature and displayed in the resulting cluster list.

Configure Client Computer for HPC Pack

This configuring applies to all versions of HPC Pack.

Note

If you are using HPC Pack in a network share installation, the network share location must be in the “Intranet” zone. You might need to adjust the Internet Options for your cluster nodes and add the network share location to the list of Intranet sites.

  1. Open a command window with administrator privileges and run the following file command

    matlabroot\toolbox\parallel\bin\MicrosoftHPCServerSetup.bat -client 

    This command performs some of the setup required for a client machine.

    Note

    If you need to override the default values the script, modify the values defined in MicrosoftHPCServerSetup.xml before running MicrosoftHPCServerSetup.bat. Use the -def_file argument to the script when using a MicrosoftHPCServerSetup.xml file in a custom location. For example:

    MicrosoftHPCServerSetup.bat -client -def_file <filename>
  2. To submit jobs or discover the cluster from MATLAB, the Microsoft HPC Pack client utilities must be installed on your MATLAB client machine. If they are not already installed and up to date, ask your system administrator for the correct client utilities to install. The utilities are available from Microsoft download center.

    If you have installed multiple versions of the Microsoft HPC Pack client utilities, MATLAB uses the most recent install. To configure MATLAB to use a specific install, set the environment variable 'MATLAB_HPC_SERVER_HOME' to the install location of the client utilities you want to use.

Validate Installation Using Microsoft HPC Pack

This procedure verifies that your parallel computing products are installed and configured correctly for using Microsoft Windows HPC Pack or Compute Cluster Server (CCS).

Step 1: Create a Cluster Profile

In this step you create a cluster profile to use in subsequent steps. To create a cluster profile, try discovering your cluster. On the Home tab, in the Environment area, select Parallel > Discover Clusters. For more information, see Discover Clusters (Parallel Computing Toolbox). Alternatively, if your cluster is not available for discovery, follow these steps.

  1. Start the Cluster Profile Manager. On the Home tab, in the Environment area, select Parallel > Create and Manage Clusters.

  2. Create a new profile in the Cluster Profile Manager by selecting Add Cluster Profile > HPC Server.

  3. With the new profile selected in the list, click Rename and edit the profile name to be HPCtest. Press Enter.

  4. In the Properties tab, provide text for the following fields:

    1. Set the Description field to For testing installation with HPC Server.

    2. Set the NumWorkers field to the number of workers you want to run the validation tests on, within the limitation of your licensing.

    3. Set the Host field to the name of the host on which your scheduler is running. Depending on your network, this might be a simple host name, or it might have to be a fully qualified domain name.

      Note: The following four property settings (JobStorageLocation, ClusterMatlabRoot, ClusterVersion, and UseSOAJobSubmission) are optional, and need to be set in the profile here only if you did not run MicrosoftHPCServerSetup.bat as described in Configure Cluster for Microsoft HPC Pack, or if you want to override the setting established by that script.

    4. Set the JobStorageLocation to the location where you want job and task data to be stored. This must be accessible to all the worker machines.

      Note

      JobStorageLocation should not be shared by parallel computing products running different versions; each version on your cluster should have its own JobStorageLocation.

    5. Set the ClusterMatlabRoot to the installation location of the MATLAB to be executed by the worker machines, as determined in Chapter 1 of the installation instructions.

    6. Set the ClusterVersion field to HPCServer or CCS.

    7. If you want to test SOA job submissions on an HPC Server cluster, set UseSOAJobSubmission to true. If you plan to use SOA job submissions with your cluster, you should test this first without SOA submission, then later return and test it with SOA job submission. The default value is determined at runtime based on your scheduler.

      So far, the dialog box should look like the following figure:

      HPCtest cluster with properties

  5. Click Done to save your cluster profile.

Step 2: Validate the Configuration

In this step you validate your cluster profile, and thereby your installation. You can specify the number of workers to use when validating your profile. If you do not specify the number of workers in the Validation tab, then the validation will attempt to use as many workers as the value specified by the NumWorkers property on the Properties tab. You can specify a smaller number of workers to validate your configuration without occupying the whole cluster.

  1. If it is not already open, start the Cluster Profile Manager from the MATLAB desktop by selecting on the Home tab in the Environment area Parallel > Create and Manage Clusters.

  2. Select your cluster profile in the listing.

  3. Click Validation tab.

  4. Use the checkboxes to choose all tests, or a subset of the validation stages, and specify the number of workers to use when validating your profile.

  5. Click Validate.

The Validation Results tab shows the output. The following figure shows the results of a profile that passed all validation tests.

Cluster Profile Manager with the local cluster profile selected. The validation results for the local profile are shown in the right panel

Note

If your validation does not pass, contact the MathWorks install support team.

If your validation passed, you now have a valid profile that you can use in other parallel applications. You can make any modifications to your profile appropriate for your applications, such as NumWorkersRange, AttachedFiles, AdditionalPaths, etc.

To save your profile for other users, select the profile and click Export, then save your profile to a file in a convenient location. Later, when running the Cluster Profile Manager, other users can import your profile by clicking Import.