https://community.cloudera.com/t5/Support-Questions/How-to-get-the-non-group-by-columns-in-spark-structured/m-p/200512#M162533 <P><A rel="user" href="https://community.cloudera.com/users/24259/elangork.html" nodeid="24259">@elango vaithiyanathan</A> </P><P>Apply <STRONG>Aggregate(agg)</STRONG> function after group by</P><PRE>val grpbyDF = updatedDf.groupBy("ROW_ID","ODS_WII_VERB").agg(max("STG_LOAD_TS"))</PRE><P>Now the <STRONG>grpbyDF</STRONG> dataframe is going to group by <STRONG>ROW_ID,ODS_WII_VERB</STRONG> and gets max value of <STRONG>STG_LOAD_TS</STRONG> column.</P> Sat, 03 Feb 2018 22:47:43 GMT Shu_ashu 2018-02-03T22:47:43Z