Group by clause in CDS view for large data set causing performance issues

We are trying to build a custom CDS view on CDS view P_ACDOCA_CUBE where we are trying to use Group by clause on certain fields. However, it seems to be causing a performance issue. Run time is increased for the same data set by a couple of minutes i.e. it's about 45 sec without group by & increases to 4-5 minutes with the group by clause.

Just to give some sense of the data volume it's like 1.5 million records.

Not sure if anyone has faced a similar issue and what would you recommend?