GroupBy node processing slowly

I am grouping a dataset,with 230,000 rows and five columns, with three columns for grouping and only one column for aggregation (unique count).

• Using an R script takes less than 3 minutes.

• On KNIME, it may take over 20 minutes.

• I have changed the memory setting to -Xmx28672m in knime.ini.

• I have tried processing in memory, but I get an error: [execute failed: java heap space]