Hive non-interactive mode and hive Mode

Hive non-interactive mode and hive Mode

Run $ HIVE_HOME/bin/hive on a linux terminal to enter the interaction mode;

$ HIVE_HOME/bin/hive-e or-f is in non-interactive mode.

1. Execute HQL statements in non-interactive mode

$HIVE_HOME/bin/hive  -e 'select * from movie limit 10'

The progress of mapreduce is displayed, but the hive interaction mode is not displayed. The final query result is

2. Execute HQL statements in non-interactive mode (-S mute mode)

$HIVE_HOME/bin/hive -S -e 'select * from movie limit 10'

Adding the-S option does not display the mapreduce progress

[hadoop@gaoyuan hive-0.14]$ bin/hive -S -e 'select * from movie limit 10'SLF4J: Class path contains multiple SLF4J bindings.SLF4J: Found binding in [jar:file:/home/hadoop/hadoop-2.4.0/share/hadoop/common/lib/slf4j-log4j12-1.7.5.jar!/org/slf4j/impl/StaticLoggerBinder.class]SLF4J: Found binding in [jar:file:/home/hadoop/hive-0.14/lib/hive-jdbc-0.14.0-SNAPSHOT-standalone.jar!/org/slf4j/impl/StaticLoggerBinder.class]SLF4J: See http://www.slf4j.org/codes.html#multiple_bindings for an explanation.SLF4J: Actual binding is of type [org.slf4j.impl.Log4jLoggerFactory]Error rolling back: Can't call rollback when autocommit=true1113382172063197110411095751161610611978771214588122118913751710140623

3. Export hive query to a file

$HIVE_HOME/bin/hive -S -e 'select * from movie limit 10' > /gaoyuan/ouput.txt

This will import the execution result of the hivequery statement to the output.txt file.

4. Run the hive script instead of entering hive interaction mode.

$HIVE_HOME/bin/hive -f hiveScript.sql    

You can also save the output result to a file.

$ HIVE_HOME/bin/hive-f hiveScript. SQL>/home/username/output.txt

 

The differences between Interactive shell and non-interactive shell

In interactive mode, shell waits for your input and runs the command you submit. This mode is called interactive because shell interacts with users. This mode is also very familiar to most users: logon, command execution, and logout. When you sign back, shell is terminated.
Shell can also run in another mode: non-interactive mode. In this mode, shell does not interact with you, But reads commands stored in files and runs them. When it reads the end of the file, shell is terminated.

What is Hive?

In addition, hive supports developers familiar with map-reduce to use the map-reduce program to perform more complex data analysis. Hive can combine thrift with control delimiters, and supports custom delimiters. Hive is based on hadoop. hadoop is a batch processing system and cannot store low latency. Therefore, hive queries cannot guarantee low latency. Hive works in the following mode: Submit a task and be notified when the task ends, instead of querying in real time. Correspondingly, a system like Oracle responds very quickly when it runs on a small data set. However, it may take several hours to process a large data set. It should be noted that hive may take several minutes to complete even if it runs on a small dataset. Low latency is not the primary goal of hive.