sql server select first row from a group

by 
select a,b
from (
select a,b,row_number() over(partition by a order by b desc) as roworder
from myTable
) temp
where roworder = 1

see http://msdn.microsoft.com/en-us/library/ms186734.aspx

More Related Contents:

  1. Add a column with a default value to an existing table in SQL Server
  2. NULL values inside NOT IN clause
  3. What is the best way to paginate results in SQL Server
  4. SQL MAX of multiple columns?
  5. Table Naming Dilemma: Singular vs. Plural Names [closed]
  6. Simplest way to do a recursive self-join?
  7. How can I convert bigint (UNIX timestamp) to datetime in SQL Server?
  8. How to calculate percentage with a SQL statement
  9. A table name as a variable
  10. How do you return the column names of a table?
  11. How to set table name in dynamic SQL query?
  12. TSQL PIVOT MULTIPLE COLUMNS
  13. INT vs Unique-Identifier for ID field in database
  14. SQL Server date format function
  15. Composite Primary key vs additional “ID” column?
  16. Fastest way to determine if record exists
  17. SQL – Query to get server’s IP address
  18. How to concatenate many rows with same id in sql?
  19. How do I edit a table in order to enable CASCADE DELETE?
  20. How to select date without time in SQL
  21. Generate nested nth level JSON SQL Server using recursive CTE
  22. Update query using Subquery in Sql Server
  23. “SELECT DISTINCT” ignores different cases
  24. T-SQL: checking for email format
  25. Database Naming Conventions by Microsoft?
  26. Functions vs Stored Procedures
  27. Convert date to YYYYMM format
  28. Combine Two Tables in Select (SQL Server 2008)
  29. T-SQL Conditional Order By
  30. Is there an Oracle equivalent to SQL Server’s OUTPUT INSERTED.*?

Leave a Comment