How is testing *system test and uat done in big data environments

ARe there separate test clusters .i.e separate name node… data nodes… how is spark and scala testing done… ? or lets say there is a map reduce code… how will test team test it ? I am just thinking out loud here…as I have not worked actually in big data space

On similar lines other question is how is debugging on a scala script done ?

There are no sophisticated tools for unit testing for Data processing.

ARe there separate test clusters .i.e separate name node… data nodes… how is spark and scala testing done… ? or lets say there is a map reduce code… how will test team test it ?

1st method: Have separate clusters for development, testing and production where each environment have its own set of nodes such as namenode, datanode etc.

2nd method: Multi tenancy - By using capacity or fair scheduler one can dedicate lower percentage of resources for testing

Each of it have its own advantage and disadvantages. But both are common methods.

There will be separate QA server(clusters) instance to perform testing process.

What about MRunit tool? They are using in the industry or just available in the market.