WITH CTE (

[ColumnName]

,[DuplicateCount]

)

AS(

SELECT

[ColumnName]

,[DuplicateCount]

,ROW_NUMBER() OVER(PARTITION BY [ColumnName]

ORDER BY [ColumnName]

AS DuplicateCount

FROM sample_table

)

DELETE

FROMCTE

WHERE (DuplicateCount > 1)

I think this an efficient way.