Support for Databases in SAS^® Viya^® 3.4

Support for Amazon Redshift

SAS/ACCESS Interface to Amazon Redshift includes a required ODBC driver.

The required client software is installed automatically. In order to reference a Data Source Name (DSN) in your connections, some post-installation configuration might be required. Consult the SAS^® Viya^® 3.4: Deployment Guide for instructions.

Support for IBM DB2

IBM DB2 Connect™ must also be licensed if you plan to connect to IBM DB2 databases that are running on AS/400, VSE, VM, MVS, and z/OS systems.

The following DBMS products are supported:

• IBM DB2 version 10.5 or later
• Client utilities for IBM DB2 version 10.5 or later

SAS recommends installing the latest FixPack on the client and server.

You must install the client software on the CAS controller and on any CAS workers that will participate in a multinode data transfer request. Multinode is an optional data transfer method that increases the speed of serial data transfers.

Support for Google BigQuery

Base SAS is required for the installation of SAS/ACCESS Interface to Google BigQuery.

The required Google BigQuery client library is included with SAS/ACCESS Interface to Google BigQuery.

Support for Greenplum

SAS/ACCESS Interface to Greenplum includes the required 64-bit ODBC driver.  Greenplum Database version 4.3 or later is supported.

Support for Hadoop

Details about the minimum supported versions for Hadoop distributions and Kerberos are provided in the following table.
Unless noted otherwise:

• SAS software listed below is for the Viya 3.4 release with the latest hotfixes applied.
  
• The HADOOPPLATFORM=SPARK option requires Spark 2. This option is available for use with SAS In-Database Technologies.

Read SAS Support for Alternative Releases of Hadoop Distributions to understand SAS support for later versions of Hadoop distributions.
Information about Hadoop JAR files and SAS environment variables for Hadoop is provided in the Hadoop Deployment: Configuring SAS Access to Hadoop and SAS Data Connector to Hadoop.

SAS In-Database Technologies for Hadoop requires the following components:

• Apache Hive
  Hive requires the following additional software:
  • Oracle JRE version 1.8 or a later version
  • Python, strace, and wget (which are included with Linux)
• MapReduce
• YARN
• HCatalog for processing non-delimited Hive file types
• SAS Embedded Process for Hadoop

Note: In order to load data in parallel with the SAS Embedded Process, the CAS controller and each CAS worker must have an IP address to which the SAS Embedded Process nodes can be externally routed.

Installation of the SAS Embedded Process with Ambari requires the free GNU wget utility. Before you use Ambari to deploy a SAS In-Database Technologies product, verify that wget is installed on every target node.

SAS identifies the specific set of Hadoop distributions that are supported with each SAS product release. The SAS policy that applies to alternative releases or distributions of Hadoop is documented at https://support.sas.com/en/documentation/third-party-software-reference/9-4/support-for-hadoop.html. The same policy that applies to SAS 9.4 also applies to SAS Viya.

Cloud Variants of Hadoop
Amazon Elastic MapReduce is supported by both SAS/ACCESS Interface to Hadoop and SAS In-Database Technologies for Hadoop.

Support for HAWQ

SAS/ACCESS Interface to HAWQ includes the required 64-bit ODBC driver.  Apache HAWQ Database version 2.2 or later is supported.

Support for Impala

SAS/ACCESS Interface to Impala supports Impala Server version 2.5 or a later version. It also supports the ODBC Driver for Impala, version 2.5.34 or a later version.

In addition, the ODBC Driver for Impala requires a compatible ODBC Driver Manager, such as the Unix ODBC Driver Manager.

You must install the client software on the CAS controller and on any CAS workers that will participate in a multinode data transfer request. Multinode is an optional data transfer method that increases the speed of serial data transfers.

Requirements for JDBC Access

A JDBC driver is required for the data source from which you want to access data. JDBC drivers are available from DBMS vendors and other third-party JDBC driver developers.

You must install the JDBC driver on the CAS controller and on any CAS workers that will participate in a multinode data transfer request. Multinode is an optional data transfer method that increases the speed of serial data transfers.

Support for Microsoft SQL Server

SAS/ACCESS Interface to Microsoft SQL Server supports Microsoft SQL Server 2012 or later.

The client software is installed automatically along with SAS/ACCESS Interface to Microsoft SQL Server. Some post-installation configuration might be required for the client.

You must install the client software on the CAS controller and on any CAS workers that will participate in a multinode data transfer request. Multinode is an optional data transfer method that increases the speed of serial data transfers.

SAS/ACCESS Interface to Microsoft SQL Server supports the following cloud variants of Microsoft SQL Server:

• Amazon RDS Microsoft SQL Server (Microsoft SQL Server 2012 or later)
• Microsoft Azure SQL Database
• Microsoft Azure SQL Database managed instance

Support for MySQL

SAS/ACCESS Interface to MySQL supports the following DBMS products:

• MySQL Client version 5.7 or later
• MySQL Server version 5.7 or later

You must install the client software on the CAS controller and on any CAS workers that will participate in a multinode data transfer request. Multinode is an optional data transfer method that increases the speed of serial data transfers.

SAS/ACCESS Interface to MySQL supports the following cloud and database variants of MySQL:

• Amazon Aurora (MySQL engine version 5.7 or later)
• Amazon RDS MariaDB (engine version 10.1 or later)
• Amazon RDS MySQL (engine version 5.7 or later)
• Azure Database for MySQL (engine version 5.7 or later)
• Google Cloud Platform Cloud SQL MySQL (engine version 5.7 or later)
• MariaDB 10.1 or later
• MemSQL 6.0 or later¹

¹SAS FedSQL and DS2 are not supported with MemSQL

Support for Netezza

SAS/ACCESS Interface to Netezza requires an ODBC driver from IBM. To obtain the appropriate IBM Netezza ODBC driver, contact IBM Technical Support at (877) 426-6006 or visit the IBM Fix Central web site: http://www.ibm.com/support/fixcentral.

SAS Viya supports the following DBMS products:

• IBM Netezza version 7.0 or later         
• Client utilities for IBM Netezza version 7.0 or later

For best results, match the Netezza ODBC client version with the version of the Netezza server where it will be connected. For example, if you have IBM Netezza Interface server release 7.0.4, you should use the ODBC client driver release 7.0.4 with SAS/ACCESS Interface to Netezza.

Support for ODBC

Before you can use SAS Viya with ODBC, an ODBC driver is required for the data source from which you want to access data. ODBC drivers are often available from DBMS vendors and other third-party ODBC driver developers. Your ODBC driver must comply with the ODBC 3.5 (or later) specification.

With SAS Viya on Windows, install the ODBC driver on the machine with SAS Viya. With SAS Viya on Linux, you must install the ODBC driver on the CAS controller, on any CAS workers that will participate in a multinode transfer request, and on the SAS Studio machine. Multinode is an optional data transfer method that increases the speed of serial data transfers. It is only supported by SAS Viya on Linux.

Note: The ODBC driver that you select might require additional DBMS software in order to enable network access.

Support for Oracle

SAS Viya requires the following Oracle components:

• Oracle Database 19c or later
• Oracle Client 19c or later (64-bit libraries)

You must install the client software on the CAS controller and on any CAS workers that will participate in a multinode data transfer request. Multinode is an optional data transfer method that increases the speed of serial data transfers.

SAS/ACCESS Interface to Oracle supports the following cloud variants of Oracle:

• Amazon RDS Oracle (19c or later)
• Oracle Cloud Platform (19c or later)

Support for PostgreSQL

SAS Viya can connect to PostgreSQL Database version 9.4.4 or a later version.

SAS Viya requires a driver manager and an ODBC driver for PostgreSQL. SAS provides both of these ODBC client components and installs them automatically.

In order to reference a Data Source Name (DSN) in your connections, some post-installation configuration might be required. Consult the SAS Viya 3.4: Deployment Guide for instructions.

With SAS Viya on Linux, you must install the client software on the CAS controller and on any CAS workers that will participate in a multinode data transfer request. Multinode is an optional data transfer method that increases the speed of serial data transfers. It is only supported by SAS Viya on Linux.

SAS/ACCESS Interface to PostgreSQL supports the following cloud variants of PostgreSQL:

• Amazon Aurora (PostgreSQL engine version 9.6 or later)
• Amazon RDS PostgreSQL (engine version 9.6 or later)
• Azure Database for PostgreSQL (engine version 9.6 or later)
• Google Cloud Platform Cloud SQL for PostgreSQL (engine version 9.6 or later)

Support for SAP ASE

SAS/ACCESS Interface to SAP ASE (on SAS Viya) requires SAP ASE (formerly Sybase) Open Client SDK, Release 15.7 or later (64-bit libraries).

Support for SAP HANA

SAS/ACCESS Interface to SAP HANA requires the ODBC driver (64-bit) for SAP HANA from SAP. This driver is part of the SAP HANA Client.

The following SAP products are required:

• SAP HANA SPS 11 Server or later
• SAP HANA ODBC Client for SPS 11 or later

You must install the client software on the CAS controller and on any CAS workers that will participate in a multinode data transfer request. Multinode is an optional data transfer method that increases the speed of serial data transfers.

Support for SAP R/3

SAS Viya requires the following products:

• SAP NetWeaver 7.0 or later
• SAP NetWeaver RFC library, Release 7.20 or later

Support for Snowflake

SAS/ACCESS Interface to Snowflake (on SAS Viya) requires the following products:

• ODBC Driver for Snowflake, version 2.19.2 or later
• The unixODBC Driver Manager. Use the latest version.

Support for Spark

SAS/ACCESS Interface to Spark and SAS In-Database Technologies for Spark work with SAS Viya 3.4 to access Apache Spark data from a Hadoop cluster.

Apache Spark 2.1 or later is required. It must be installed on the Hadoop nodes.

For Apache Spark support, SAS Viya also requires one of the following Hadoop third-party distributions:

• Cloudera CDH 5.5 and later releases (required if using SAS In-Database Technologies for Hadoop to access Spark data)
• Hortonworks HDP 2.6 and later releases (required if using SAS/ACCESS Interface to Spark or SAS In-Database Technologies for Spark)

SAS^® In-Database Technologies for Spark requires the following components to be installed on the Spark cluster:

• Apache Hive
  Hive requires the following additional software:
  • Oracle JRE version 1.8 or a later version
  • Python, strace, and wget (which are included with Linux)
• MapReduce
• YARN
• HCatalog for processing non-delimited Hive file types.
• SAS Embedded Process for Spark, which is included with SAS In-Database Technologies for Spark

In order to load data in parallel with the SAS Embedded Process, the CAS controller and each CAS worker must have an IP address to which the SAS Embedded Process nodes can be externally routed. Installation of the SAS Embedded Process with Ambari requires the free GNU wget utility. Before you deploy a SAS In-Database Technologies product with Ambari, verify that wget is installed on every target node.

SAS/ACCESS^® Interface to Spark requires Hortonworks HDP 2.6 or later. In addition, Apache Spark 2.1 or later must be installed on the Hadoop nodes. When the playbook detects the presence of Spark, the remaining configuration is performed automatically.

Support for Teradata

SAS Viya support for Teradata requires the following products:

• Teradata Database version 15.10 or later
• Teradata CLIv2 client libraries, TTU 15.10 or later for Linux (64-bit libraries)*

*Note: SAS/ACCESS Interface to Teradata offers limited support for TTU 17 and later. SAS recommends that you use TTU 16.20 instead. For more information, see SAS Note 69317.      

You must install the client software on the CAS controller and on any CAS workers that will participate in a multinode data transfer request. Multinode is an optional data transfer method that increases the speed of serial data transfers.

The SAS Data Connect Accelerator for Teradata also requires SAS Embedded Process for Teradata.

The SAS Embedded Process for Teradata is included with SAS In-Database Technologies for Teradata. It requires 600 MB of disk space in the /opt file system on each Teradata Trusted Parallel Application (TPA) host.

In order to load data in parallel, the CAS controller and each CAS worker must have an IP address that can be routed to externally from the SAS Embedded Process TPA nodes.

Additional disk space in the /opt file system is also required on each TPA node in order to accommodate SAS Quality Knowledge Base (QKB). The size of the QKB varies, but the default size is 8 GB.

Both SAS/ACCESS Interface to Teradata and SAS In-Database Technologies for Teradata support Teradata Database 15.10 or later on the following cloud platforms:

• Teradata Cloud
• Amazon Web Services
• Microsoft Azure
• VMware

Support for Vertica

SAS/ACCESS Interface to Vertica requires Vertica Analytic Database version 7.1 or later and Vertica ODBC Client version 7.1 or later from Hewlett Packard. SAS/ACCESS Interface to Vertica also requires the Vertica ODBC Client Driver.

To obtain the Vertica Client ODBC driver, contact your database administrator or visit the myVertica website: https://my.vertica.com/download/vertica/client-drivers.

You must install the client software on the CAS controller and on any CAS workers that will participate in a multinode data transfer request. Multinode is an optional data transfer method that increases the speed of serial data transfers.

Support for Databases in SAS Visual Investigator

SAS Visual Investigator 10.4 and SAS Intelligence and Investigation Management 1.2 support the following databases:

• IBM DB2
• Microsoft SQL Server
• MySQL
• Oracle 12c
• PostgreSQL
• Teradata

Data from MySQL and Microsoft SQL Server data sources is accessible using SAS/ACCESS Interface to ODBC.