Option 1: For Multiple-Node Data Sets - Aster Analytics

Version 1.0

SELECT * FROM DistnmatchReduce (
  ON DistnmatchMultipleInput (
    ON (SELECT RANK() OVER (PARTITION BY col [,...] 
      ORDER BY column_name) AS rank, *
      FROM input_table 
      WHERE column_name IS NOT NULL
    ) AS input PARTITION BY ANY
    ON (SELECT col[,...], COUNT(*) AS group_size
      FROM input_table 
      WHERE column_name IS NOT NULL
      GROUP BY col[,...] 
    ) AS groupstats DIMENSION
  ValueColumn (column_name)
  [ Tests ('test' [,...]) ]
  Distributions ('distribution:parameter' [,...])
  [ GroupingColumns (col[,...]) ]
  [ MinGroupSize (minGroupSize) ]
  [ NumCell (cell_size) ]
  )
  PARTITION BY col[,...] 
);