Contingency environment

Hello, 

I'm interested to know what solutions should be usual for a Vertica contingency environment. 
The idea would be have two separated clusters (similar environments with the same loaded information), in active-active mode. 

The simplest way could be load both clusters via ETL processes, however it could add aditional problems and complexity to sure both are similars. 

Is there another usual solution to load data in this kind of scenario?

Regards.
Verónica