Hive Query Language
Databases in Hive:
CREATE DATA BASE db name to create database in Hive
USE db name To use the database in Hive.
DROP db name To delete the database in Hive.
SHOW DATA BASE to see the list of the DataBase
If no database is specified, tables belong to the default Data Base.
Tables in Hive:
Hive table is logically made up of the data being stored and the associated metadata describing the layout of the data in the table.
The data typically resides in HDFS, although it may reside on any Hadoop file system including the local file system.
have>CREATE TABLE EMP (empid int, ename string, esal double) ROW FORMAT DELIMITED FIELDS TERMINATED By ‘t’ LINES TERMINATED by ‘n’ STORED AS TEXT FILE;
1. Managed tables
Managed tables are the one which will be managed in the Hive warehouse i.e. whenever we create a managed table definition, it will be stored under the default location of the Hive warehouse i.e./user/Hive/ware house.
When we drop a managed table, Hive deletes the data in the table
Managed tables are less convenient for sharing with other tools.
Syntax for creating Hive managed table:-
Hive>create table manage- tab (empid, ename string, esal int) row format delimited fields terminated by ‘t’ lines terminated by ‘m’ stored as a text file;
#hadoop fs –ls/user/Hive/warehouse.
We can load the data in two ways
In local mode, the syntax is
hive>load data local in path’/home/new Batch/input1.txt’ Into table managed-tab;
For HDFS mode, the syntax is
hive>load data in path’/user/ramesh/Hive/input2.txt’
Into table managed – tab;
Once the successful loading of the table and once the file is loaded, the file will be deleted in HDFS path and we can see in use/Hive/ware house
2) External Tables:-
Along with the managed tables, Hive also uses external tables.
Whenever the key word ‘external’ comes in the table definition part. Hive will not bother about the table definition, i.e. the external table will not be managed by the Hive warehouse system.
Along with the external keyword, we can also mention the ‘location’ in the table definition, where exactly the table definition will get stored.
When you drop an external table, Hive leave the data untouched and only delete the meta data.
Hive>create external table external- tab(empid int, ename string, esal double) row format delimited fields Terminated by ‘f’ lines terminated by ‘n’ stored as text file location ‘userRameshHive-external’;
?Location will be automatically created.
Loading data into External Tables:-
Hive>load data in path’/Ramesh/input data.txt’ into table external-tab;
If we delete the managed table, both the schema and the data file will be deleted.
But, if we delete external tables, only the schema will be deleted and data file will be there in the specified location.
Difference between managed tables & External Tables:-
One of the main differences between managed and external tables in Hive is that when an external table is dropped, the data associated with it does not get deleted from only the meta data (no. of cols, types of cols, terminators etc.) gets dropped form the Hive meta store
I have so far always preferred making table external because if the schema of my Hive table changes, I can just drop the external table and recreate another external table over the same HDFS data with the new schema
Hive>Create external table log in for tab(log id int, log Error string,Log error count int) row format delimited fields terminated by’f’ stored as text file location ‘user/external location’; Hive>select*from log in for tab;
We get the result from the file which we specified in the location path
Renaming a Table:
Cmd: ALTER TABLE log _ messages RENAME To logmsgs;
You can rename a column, change its position, type or comment.
ALTER TABLE log-messages CHANCE COLUMN hms hours-minutes-
Seconds INT COMMENT ’The hours, minutes and seconds are part of the times tamp’ AFTER Severity;
You can add new columns to the end of the existing columns, before any partition .
Example: ALTER TABLE Log-message ADD COLUMNS(app-name String COMMENT” Application Name” ,session-id long);
Deleting or replacing columns:
The replace statement can only be used with tables that use one of the native ser De modules are Dynamic Ser De or Meta data Type column set ser De.
Ex:-ALTER TABLE log- messages REPLACE COLUMNS( Hours-mins-sees INT Severity STRING Message String);
?This statement effectively renames the original hms column and removes the server and process – id columns from the original schema definition.
As for all the ALTER Statements, only the table metadata is changed.
Partitioning and Bucketing:
Hive organizes tables into partitions, a way of dividing a table into course – grained parts based on the value of a partition column, such as date.
hive>create table party table(loaded int, log error string) PARTITIONED BY (Logdt string, country string) row format delimited field terminated by ‘t’ lines terminated by ‘n’ stored as text file
There are two reasons why you might want to organize your tables (or partitions) into buckets.
1. The first is to enable more efficient queries.
2. The second reason to bucket a table is to make sampling more efficient.
Tell Hive that a table should be bucketed. And we use the CLUSTERED By clause to specify the columns to bucket and the number of buckets
hive>CREATE TABLE bucketed users(id INT, name STRINA) CLUSTERED BY (id)INTO 4 BUCKETS;
Here we are using the user ID to determine the bucket the Hive does which is done by hashing the value and reducing module and the number of buckets, so any particular bucket will effectively have a random set of users in it.
The data within a bucket may additionally be stored by one or more columns.
This allows even more efficient map-side joins, since the join of each bucket becomes an efficient merge sort.
Syntax for delving that a table has sorted buckets is:
have>CREATE TABLE bucketed users(id INT, name STRING) CLUSTERED By(id)SORTED By(id ASC)INTO 4 BUCKETS;