Big Data SQL 3.2 has been released and is now available for download on edelivery. This new release has many exciting new features – with a focus on simpler install and configuration, support for new data sources, enhanced security and improved performance.
Big Data SQL has expanded its data source support to now include querying data streams – specifically Kafka topics:
This enables streaming data to be joined with dimensions and facts in Oracle Database or HDFS. It’s never been easier to combine data from streams, Hadoop and Oracle Database.
New security capabilities enable Big Data SQL to automatically leverage underlying authorization rules on source data (i.e. ACLs on HDFS data) and then augment that with Oracle’s advanced security policies. In addition, to prevent impersonation, Oracle Database servers now authenticate against Big Data SQL Server cells. Finally, secure Big Data SQL installations have become much easier to set up; Kerberos ticket renewals are now automatically configured.
There has been significant performance improvements as well. Oracle now provides its own optimized Parquet driver which delivers a significant performance boost – both in terms of speed and the ability to query many columns. Support for CLOBs is also now available – which facilitates efficient processing of large JSON and XML data documents.
Finally, there has been significant enhancements to the out-of-box experience. The installation process has been simplified, streamlined and made much more robust.
Big Data SQL requires:
Oracle Database 184.108.40.206
August 2017 Proactive BP (220.127.116.11.170814 #26610322)
One-off Big Data SQL Patch (#26170659)
CDH or HDP Cluster