Select correct statements about Hadoop MapReduce.

Reducers input key$/$value pairs is grouped by the key.

The output type of keys$/$values of mappers$/$reducers must be of the same type as their input

It is possible to start reducers while some mappers are still running.

Each mapper$/$reducer must generate the same number of output key$/$value pairs as it receives or input.

Mappers and reducers may generate any number of key$/$value pairs $($including zero$).$

Mapper may produce key$/$value pairs of any type.

The input to reducers is grouped by key