Skip to content

Latest commit

 

History

History
10 lines (7 loc) · 693 Bytes

README.md

File metadata and controls

10 lines (7 loc) · 693 Bytes

Data Open

Overview

Data Open is a web-based collaborative data analysis platform. Users can extract, analyze and visualize data by writing Spark application code and submitting it to a Hadoop cluster. Analysis results can be shared with others as a form of a notebook.

Secure

Data Open adopts LDAP authentication and authorization. A Data Open user is set to a Spark application user that works well with a Hadoop cluster whose data access is controlled via Apache Ranger. In addition, editors and readers of an analysis result can be restricted to some users and groups.

Scalable

Data Open can run on multiple servers so that it can deal with increase of the number of users.