SAS Visual Analytics and SAS Visual Statistics are designed to work together, but are separately licensed.
SAS Visual Analytics and SAS Visual Statistics are available for both distributed and non-distributed SAS LASR Analytic Server deployments. These system requirements apply to both LASR deployments. The headings below indicate whether components apply to a specific LASR deployment.
The SAS Metadata Server and the SAS Web Infrastructure Platform can be installed on machines that use a different operating system family.
Note: If you deploy the middle tier and the server tier on dissimilar operating systems (for example, one on Linux and the other on Windows), the audit and relationship data extraction and reporting features are not supported.
In a Distributed LASR deployment, SAS Visual Analytics and SAS Visual Statistics require a minimum of a 4-server, 16-core environment (4 servers X 4 cores) on commodity hardware with Intel Xeon architecture. A minimum of 16 GB of RAM is required per core (256 GB total).
In a Non-distributed LASR deployment, SAS Visual Analytics and SAS Visual Statistics require a server with a minimum of 4 cores (commodity hardware with Intel Xeon architecture). A minimum of 16 GB of RAM is required per core (64 GB total).
If the SAS Metadata Server is installed on a separate machine, see the guidelines for the Metadata Server below.
We strongly suggest working with your SAS Account Team to obtain an official hardware recommendation based on your proposed SAS workload.
SAS recommends installing the SAS Visual Analytics Server on a dedicated machine.
SAS Visual Analytics Server includes a standard SAS/ACCESS engine of choice, or you can substitute SAS Data Surveyor for SAP.
Teradata and EMC Greenplum must use the SAS Embedded Process to achieve parallel loading. To use the SAS Embedded Process, you must install the appropriate SAS/ACCESS engine on the server tier.
SAS Information Retrieval Studio 1.53 requires Python version 2.6 through 2.7.x (not 3.0 or later) for deployment. For more information, see "SAS 9.4 Support for Additional Products."
If ArcGIS for Server is used as your map service, SAS Visual Analytics requires release 10.1 or later.
The High-Performance Computing Management Console subcomponent is deployed only on a Red Hat Enterprise Linux computing grid.
Co-located Data Storage
SAS LASR Analytic Server supports Hadoop data storage environments.
Use of a co-located Hadoop File System (HDFS) is optional, but it requires a supported Hadoop environment and SAS Plug-ins for Hadoop (which is included with your license). For information about supported Hadoop distributions, refer to http://support.sas.com/resources/thirdpartysupport/v94/hadoop/hadoop-distributions.html. The plug-ins require additional configuration. For more information, see SAS High-Performance Analytics Infrastructure 3.5: Installation and Configuration Guide at http://support.sas.com/documentation/solutions/hpainfrastructure/.
The following options are also supported, with procurement and installation by the customer:
Co-located storage for Teradata and Greenplum is no longer supported for new installations. To load data in parallel from these providers, use the SAS Embedded Process. To use the SAS Embedded Process, you must install the appropriate SAS/ACCESS engine on the server tier.
Apache Hadoop software is distributed with SAS software as a convenience. For support information, see http://support.sas.com/resources/thirdpartysupport/ApacheHadoopSupport.html.
Note: Base SAS, SAS/ACCESS, and SAS Visual Analytics can either be installed on the head node of the commodity hardware (symmetric) or on separate hardware from the commodity hardware (asymmetric). This architecture enables you to install the SAS Visual Analytics, Distributed LASR server-tier Base SAS components on Windows for x64 while working with the Linux for x64 LASR grid.
SAS makes the following recommendations for SAS Metadata Server deployments:
The disk I/O performance of any server-class system should be sufficient. Memory requirements are highly dependent on the product mix and usage, but 8 GB of RAM are sufficient for most large deployments. For many deployments, 4 GB of RAM are sufficient and should be considered a minimum, unless the specific deployment is tested and verified to require less memory.
SAS Visual Statistics is always installed on the same machine as SAS Visual Analytics.
SAS recommends installing SAS Visual Analytics Mid-Tier on a dedicated machine.
When you install your SAS solution software, SAS Deployment Manager installs and configures the required web application servers and web servers.
To install these servers, select one of the supported operating systems for the middle tier.
Note: These servers cannot be installed on the z/OS or 32-bit Windows platforms.
SAS Visual Analytics Mobile Clients (SAS Mobile BI) are available for Apple and Android devices via free downloads from app stores.
The list of Web browsers that SAS supports is updated as new products are tested and approved.
The “classic” appearance of the SAS Visual Analytics user interface is supported by all of the browsers that are listed on the SAS 9.4 Support site at http://support.sas.com/resources/thirdpartysupport/v94/browsers.html.
Note: The classic appearance also requires Adobe Flash Player 11.1 or later.
The SAS Visual Analytics Viewer and the SAS Home page use the “modern” appearance by default. For a list of browsers that are required for the modern appearance, refer to the system requirements for SAS Common Components 2.4 for HTML.
If you do not have access to a browser that is supported for the modern appearance:
SAS software may be provided with certain third-party software, including but not limited to open-source software, which is licensed under its applicable third-party software license agreement. A listing of this third-party software may be found at http://support.sas.com/thirdpartylicenses.