It supports the two most widely-used Hadoop distributions: Hortonworks Data Platform and Cloudera Distribution for Hadoop. It also uses Spark on YARN, making data processing faster and more stable.
Access to individual data sets and projects can be restricted for specific users and user groups, allowing Oracle Big Data Discovery 1.1 to be rolled out to large groups of analysts securely.
Language processing capabilities are expanded to support more than 80 languages.
Improvements to My Organization
The capability to process so many languages has been our biggest improvement. We were working on customer sentiment analysis for a large media company which has recently taken over a large studio in Hollywood (anonymous for confidentiality reasons). We could work on large unstructured datasets from different parts of the world on Facebook and Twitter to get better insights.
Room for Improvement
They need to work on the visualization of data. That would be a nice, useful improvement.
Use of Solution
I've been using it since March 2016.
We had no issues with the deployment.
We had to pursue several performance tuning measures to bring up the system to scale. Oracle ACS supported us in stabilizing the system.
We have had no issues scaling it for our needs.
Customer Service and Technical Support
In my experience, I'd rate customer service a 6 out of 10. Technical Support
In my experience, I'd rate technical support a 6 out of 10.
Oracle provides organized documentation on initial setup. We engaged Oracle Advanced Customer Support (ACS) along with our infrastructure team. We utilized our Oracle Big Data Appliance (BDA) server.
We implemented it with our in-house team. It is very important to begin with an objective approach around the use cases. For instance, it took us around two months to arrive at a consensus on the kind of entertainment program that will attract more viewership towards our client’s upcoming movie channel. Our client was focused on only sports in the beginning.
Other Solutions Considered
We were also looking at QlikView, but it is not fit for scale, hence we switched to Oracle Big Data Appliance.