hive

data warehouse system for Hadoop

TLDR

Start Hive shell

$ hive

Execute query

$ hive -e "SELECT * FROM [table]"

Run script file

$ hive -f [script.hql]

Set configuration

$ hive --hiveconf [key=value]

Silent mode

$ hive -S -e "[query]"

Hive is a data warehouse system for Hadoop. It provides SQL-like query language (HiveQL) for querying large datasets stored in HDFS.The tool translates queries to MapReduce, Tez, or Spark jobs. It's used for data analysis and ETL on big data platforms.

PARAMETERS

-e QUERY

Execute query.

-f FILE

Execute script file.

-S, --silent

Silent mode.

--hiveconf KEY=VALUE

Set configuration.

--database DB

Use database.

-i FILE

Initialization file.

--help

Display help information.

INSTALL

brew install hive

CAVEATS

Requires Hadoop cluster. Query latency higher than RDBMS. Schema on read.

HISTORY

Apache Hive was developed at Facebook and contributed to the Apache project for SQL-based big data analytics.