What's New in SAS Data Integration Studio 4.901

Overview

The main enhancements and changes for the third maintenance release for SAS 9.4 for SAS Data Integration Studio include the following:
  • added three new transformations including Fork, Fork End, and Wait For Completion nodes
  • revised the command-line deployment tool for scheduling batch deployments
  • added two new options to the SAS LASR Analytic Loader transformation
  • added support for Hadoop (Hive), HAWQ, Impala, LASR, PI, or SASHDAT engines
  • enhanced the Loop transformation
  • added PI LIBNAME engine support
  • added the HAWQ source designer

New Fork, Fork End, and Wait for Completion Transformations

The Fork transformation allows pieces of a SAS job to be run in parallel with other pieces of SAS code. Each piece run in parallel is demarcated by the Fork and Fork End transformations. The Wait For Completion transformation uses the output from multiple Fork transformations, allows the job to wait for all or any of the processes to complete, and then creates a single output.

Revised Command-Line Deployment Tool

The revised command-line batch deployment tool enables users to batch deploy many jobs at once using a simple command-line interface. The user invokes an executable named “DeployJobs.exe” and supplies parameters to control its behavior. The BatchJobDeployment class retrieves the source code for each job, stores it on the application server specified by the user, and then calls another class, AppMethods, to handle the actual job deployment. All options are specified as arguments to the “DeployJobs” executable, so a manifest file is no longer required. The tool can also be used on platforms previously unsupported such as z/OS. See Using a Command Line to Deploy Jobs for more information.

New SAS LASR Analytic Server Loader Transformation Options

Two new options include the following:
  • Use the SASIOLA engine for loading, which triggers different load methods.
  • Update table metadata for the target tables, which generates PROC METALIB to update metadata at run time on the table metadata for the target.
See the SAS Data Integration Studio Help topic on Options tab of the SAS LASR Analytic Server Loader transformation for additional information.

Updated Support for Hadoop (Hive), HAWQ, Impala, LASR, PI, or SASHDAT Engines

The UPDATE statement is not supported by Hadoop (Hive), HAWQ, Impala, LASR, PI, or SASHDAT engines, so they cannot be used as target tables in the following transformations:
  • SCD Type 1
  • SCD Type 2
  • SQL Delete
  • SQL Insert Rows
  • SQL Update
  • Table Loader (with a PI target table)
See General Usage Notes for more information.

Enhancement to the Loop Transformation

The Loop transformation has been enhanced so that you can create a single job with two loops where the second loop is contained within the first loop.

Added PI LIBNAME Engine Support

Added support for the PI LIBNAME engine to register PI tables in the source designer, and to read and write PI tables. SeeTable Loader Notes When Using the PI System as a Target for any restrictions.

Added HAWQ Source Designer

Added support for Hadoop With Query (HAWQ) source designer that provides an SQL interface to store data natively in the Hadoop Distributed File System (HDFS). See Table Loader Notes When Using HAWQ as a Target for any restrictions.
Copyright © SAS Institute Inc. All rights reserved.