Item: Apache Hive
Rating: 5
Author: Verified User

Use Cases and Deployment Scope

Hive was once a part of our platform but it never lived up to the promise of performant SQL on HDFS and thus was only truly useful for the users who didn't have the expertise or time to write MapReduce. With the advent of Spark, Hive's time is numbered and I would not invest in learning it specifically but instead use SparkSQL which has some of the better parts of Hive under the covers along with Spark's better execution engine.

Pros and Cons

Connect BI tools to non relational data stores
Simplify writing legacy MapReduce

Speed needs to be a lot better
Concurrency is not up to snuff

Return on Investment

Helped ease BI tool connectivity to my NoSQL database

Alternatives Considered

SparkSQL, Tez and Impala

All are improvements over the Hive tooling and are very much inspired by Hive. Hive was selected before they were on the market.

Other Software Used

Cassandra, Apache Spark

Likelihood to Recommend

Hive is mostly useful in HDFS environments where legacy BI tools need to access the data. This is ok if there is a low concurrency of users but will fall over with any significant multi-user environment.

Hive, last generation tooling but revolutionary for it's time.

Overall Satisfaction with Apache Hive