Most analytics solutions on the market today can now access many types of data sources, including modern systems like Hadoop. It’s a mandatory requirement these days, because most enterprises (probably all enterprises) store useful data in all sorts of datastores throughout the organization. The popular analytics platform, Tableau, is one such tool that can combine data from disparate sources, reducing the amount of data that must be ETL’d around just to satisfy a report request. This approach has the potential for slowing performance and increasing my memory/cpu consumption on the application side, as previously noted. Not the ideal situation for a business analyst.
Our other option is to virtualize the Impala table and make it accessible from within Oracle. We can do this using the Present functionality of the Gluent Data Platform. No code to write, no ETL streams, no data movement, and no changes to your existing Tableau reports. Presenting a table from a Hadoop data source as an external table in your Oracle database is the quickest way to access Hadoop first data in real-time
Originally on gluent.com — read the full post here…