Deploying the Infrastructure

The following list summarizes the steps required to install and configure the SAS High-Performance Analytics infrastructure:

1. Create a SAS Software Depot.

2. Check for documentation updates.

3. Prepare your analytics cluster.

4. (Optional) Deploy SAS High-Performance Computing Management Console.

5. (Optional) Modify co-located Hadoop.

6. Deploy the SAS High-Performance Analytics environment.

7. (Optional) Deploy the SAS Embedded Process for Hadoop.

8. (Optional) Configure the analytics environment for a remote parallel connection

The following sections provide a brief description of each of these tasks. Subsequent chapters in the guide provide the step-by-step instructions.

Create a SAS Software Depot, which is a special file system used to deploy your SAS software. The depot contains the SAS Deployment Wizard—the program used to install and initially configure most SAS software—one or more deployment plans, a SAS installation data file, order data, and product data.

For more information, see Creating a SAS Software Depot in SAS Intelligence Platform: Installation and Configuration Guide.

It is very important to check for late-breaking installation information in SAS Notes and also to review the system requirements for your SAS software.

Preparing your analytics cluster includes tasks such as creating a list of machine names in your grid hosts file. Setting up passwordless SSH is required, as well as considering system umask settings. You must determine which operating system is required to install, configure, and run the SAS High-Performance Analytics infrastructure. Also, you will need to designate ports for the various SAS components that you are deploying.

For more information, see Preparing Your System to Deploy the SAS High-Performance Analytics Infrastructure.

SAS High-Performance Computing Management Console is an optional web application tool that eases the administrative burden on multiple machines in a distributed computing environment.

For example, when you are creating operating system accounts and passwordless SSH on all machines in the cluster or on blades across the appliance, the management console enables you to perform these tasks from one location.

For more information, see Deploying SAS High-Performance Computing Management Console.

If your site wants to use Hadoop as the co-located data store, then you can modify a supported pre-existing Hadoop distribution.

The SAS High-Performance Analytics environment consists of a root node and worker nodes. The product is installed by a self-extracting shell script.

Software for the root node is deployed on the first host. Software for a worker node is installed on each remaining machine in the cluster or database appliance.

For more information, see Deploying the SAS High-Performance Analytics Environment.

Together the SAS/ACCESS Interface and SAS Embedded Process provide a high-speed parallel connection that delivers data from the co-located SAS data source to the SAS High-Performance Analytics environment on the analytics cluster. These components are contained in a deployment package that is specific for your data source.

For information about installing the SAS Embedded Process, see the SAS In-Database Products: Administrator’s Guide.

You can optionally configure the SAS High-Performance Analytics Environment for a remote parallel connection.

For more information, see Configuring the Analytics Environment for a Remote Parallel Connection.