Objective

To install Pig on Clustering and make it run automatically on system startup.

Steps

Choose 'Pig' when install Hadoop through Ambari. Please find more detailes in here.
Set up Pig configuration file and environmen variables, you can follow the instructions here. You can also set the password(optional).
To validate whether Pig has been installed rightly, run instructions here. Execute the following commands :
- hdfs dfs -copyFromLocal /etc/passwd passwd
- Create the pig script file /tmp/id.pig with the following contents:
```
A = load 'passwd' using PigStorage(':');
B = foreach A generate $0 as id; store B into '/tmp/id.out';
```
- Execute the Pig script, run pig -l /tmp/pig.log /tmp/id.pig , for more details, you can see here
- There are some arguments you may want to use, e.g. pig -help for more details.
You can also install 'Pig' by adding the service 'Pig' through your Ambari server webpage after you install and setup well the cluster(i.e., you can run a simple mapreduce program on Hadoop). If you are informed to install Tez as well, just click yes and install it.
here provides detailed instructions and simple codes to deploy the Pig application.