Installing apache impala on the remote server without hadoop and hive

[arun kumar <aa...@gmail.com>]

 Hi Team,

I have installed apache hadoop and apache hive on  centos 7 machine(node1).

I will install apache impala on a separate node(node2)  which is isolated
from node1 and no other hadoop service running go on it.

I wanted to know below scenario is feasible or not:

1)I wanted to run an impala  job from node2  which will take the data from
hdfs (node1) and process it.

2)If this is possible ,Can impala directly read the data from hdfs(node1)
or do I need to install any other services on node2 so that node1 and node2
can communicate.

3)What should be the better solution for this scenario.

Please find the attachment for better understanding.

Thanks in advance!!

Regards
Aarun