In the fourth maintenance release for SAS^® 9.4 (TS1M4), a new LIBNAME option, SCRATCH_DB=, has been added to the SAS/ACCESS^® Interface to Hadoop. This option enables you to specify a Hive schema to which the user has Write access for writing temporary data. By default, when doing Hadoop Distributed Files System (HDFS) reads, temporary output is written into the same schema that is being read. So, this option is particularly useful when the schema being read from is a Read-only schema. You can force Java Database Connectivity (JDBC) reads by using the option READ_METHOD=JDBC, but that read method might be substantially slower than HDFS reads.

Documentation for the SCRATCH_DB options can be found in the "SCRATCH_DB LIBNAME Option" section of the SAS/ACCESS^® 9.4 for Relational Databases: Reference, Ninth Edition.

Click the Hot Fix tab in this note to access the hot fix for this issue.

Operating System and Release Information

A fix for this issue for SAS/ACCESS Interface to Hadoop 9.43 is available at: