Changes in Oracle Big Data SQL 4.1
The following are new features and updates in Oracle Big Data SQL 4.1.
Oracle SQL Access to Kafka
Oracle SQL Access to Kafka enables Oracle external tables to query data from Kafka topics. This feature allows Kafka data to be queried and joined with data in Oracle Database tables without requiring Hive metadata. See Oracle SQL Access to Kafka.
Object Store Reader Enhancements
The following ORACLE_BIGDATA driver updates are available:
- Support for Apache ORC file format. Create an Oracle External Table for Object Store Access.
- Support for complex data types with ORC, Parquet and Avro.
This capability was supported in Hadoop, and is now supported with ORACLE_BIGDATA. See Create an Oracle External Table for Object Store Access.
- Enhancements to Text Driver
New text-based access parameters available. See ORACLE_BIGDATA Access Parameters.
Support for Microsoft's Azure Blob Object Storage
Microsoft's Azure Blob Storage is supported in this release. The ORACLE_BIGDATA driver has been extended to support the creation of external tables that map to data stored in Azure Blob storage. See Create an Oracle External Table for Object Store Access.
Supports Oracle Database 19c
Note:
Sentry is not supported for Big Data SQL installations on Oracle Database 12.1 systems.- Hybrid Partitioned Tables (HPT)
See Information Lifecycle Management: Hybrid Access to Data in Oracle Database and Hadoop.
- In-Memory External Tables
See In-Memory External Tables in Oracle Database Database In-Memory Guide 19c.
Installer Improvements
- Alternate Kerberos Principals
On the database side of the Big Data SQL installation, Release 4.1 supports Kerberos authentication via the principal of a user other than the database owner. The --alternate-principal parameter is described in Command Line Parameter Reference for bds-database-install.sh in Oracle Big Data SQL Installation Guide.
- Alternate Repositories for Cloudera Enterprise 6 Release (CDH 6)
On CDH 6 systems only, you can now specify an arbitrary repository for the installer to acquire the Hadoop, Hive, and HBase clients needed to complete the Oracle Big Data SQL installation on the database side. The --alternate-repo parameter is described in Command Line Parameter Reference for bds-database-install.sh in Oracle Big Data SQL Installation Guide.
- Support for Non-Default Hive and Hadoop Owners and Groups
Release 4.1 includes new Jaguar configuration parameters to identify the correct user and/or group for HDFS and Hive daemons when these have been changed from the defaults. See hadoop_ids in Jaguar Configuration Parameter and Command Reference in Oracle Big Data SQL Installation Guide.