Search code examples
hadoopmapreducehbasebigdata

MAPREDUCE - BULK LOADING DATA INTO HBASE TABLE

Why do we use only a driver class and a mapper class and we don't use o reducer class?

Solution

  • Reducer can be used if you want to do any aggregations on the bulk loaded data.

    In normal case if you are just loading with out any aggregations. then mapper only jobs are enough.

    For example :

    case 1:

    if you are reading the CSV and loading all words in hbase table with respective columns i.e with out aggregations like wordcount. then mapper only jobs are enough.

    case 2:

    if you are reading csv and want to do aggregations like word count ,( which is aggregation) then load it in hbase then you need reducer.

    Hope that clarifies..