Apache: Big Data 2016

Sentry is an framework to provide fine-grained access control on data stored on a Hadoop cluster. Sentry has been leveraged to manage authorization policies to Hive, Solr, Impala, and Kafka. A new generic authorization model has been implemented in Sentry to allow enabling protections on various types of data existing in Hadoop engines easily. It is critical to instrument different authorization security models meeting diverse security requirements. The architecture of the new generic model is designed to plug-in various authorization model such as role based access control and attribute based access control easily using the same storage service. In this talk, Anne and Hao will present the architecture of the generic model in Sentry framework which satisfies all those targets. And they will also elaborate how to create policies to protect data for different Hadoop engines.