For many years, Oracle Database has provided rich support for Information Lifecycle Management (ILM). Numerous capabilities are available for data tiering – or storing data in different media based on access requirements and storage cost considerations. These tiers may scale from in-memory for real time data analysis – to Database Flash for frequently accessed data – to operational data captured in Database Storage and Exadata Cells.
Hadoop offers yet another storage layer for the – the Hadoop Distributed File System (HDFS) – which offers a cost effective alternative for storing massive volumes of data. Oracle Big Data SQL makes access to this data seamless from Oracle Database 12c; Big Data SQL is a data virtualization technology that allows users and applications to use Oracle’s rich SQL language across data stored in Oracle Database, Hadoop and NoSQL stores. One query can combine data from all these sources.
What this means is that ILM can now be extended to use Hadoop to store raw and archived data. This is especially important since retaining many years of historical information in data warehouses is increasingly a requirement for both analytics and regulatory compliance...