SlideShare a Scribd company logo

SQL Server2012 Enhancements

•Download as PPTX, PDF•
•
Abhishek Sur
Abhishek Sur

This document discusses new features in SQL Server 2012 including Always On, contained databases, columnstore indexes, Visual Studio integration, and TSQL enhancements. It provides details on columnstore indexes, query pagination using new features, windowing functions using the OVER clause, sequences, metadata discovery using new DMVs and stored procedures, enhanced functions, and general TSQL improvements including THROW and extended events.

1 of 15
By Abhishek Sur Microsoft MVP, Client App Dev Follow us at Twitter @abhi2434 & @sqlservergeeks
Why SQL Server 2012 ?  Always On  Contained Databases  Column Store Indexes  Visual Studio integration with Management Studio  TSQL Enhancements 2
3
The Concept RowStore Page Int Varchar Datetime float • ColumnStore reduces redundancy of data and compression as the data type remains same for the entire page. • All data for a single column could be retrieved from a non clustered Columnstore Page index page. Int Int Int Int float float float float Limitation • 1 Table can have one columnstore index. 4
5
Introducing SQL Server 2012 Transact-SQL Improvements • Query Pagination • OVER Clause Windowing • Sequences • Metadata Discovery • Enhanced Function Library • General T-SQL-Related Enhancements
Query Pagination • Common application requirement to retrieve rows − For a given page − For a given page length − Our Options ??? − Retrieve all and filter in application end − Retrieve data using Row_NUMBER function − Any other options ? − Denali gives better option for pagination
OVER Clause Windowing • Some existing queries do not optimize well • Example: Details of orders and days since previous order of each product -- Traditional approach SELECT rs.ProductKey,rs.OrderDateKey,rs.SalesOrderNumber, rs.OrderDateKey- (SELECT TOP 1 prev.OrderDateKey FROM dbo.FactResellerSales AS prev WHERE rs.ProductKey=prev.ProductKey AND prev.OrderDateKey<=rs.OrderDateKey AND prev.SalesOrderNumber<rs.SalesOrderNumber ORDER BY prev.OrderDateKey DESC, prev.SalesOrderNumber DESC) AS DaysSincePrevOrder FROM dbo.FactResellerSales AS rs ORDER BY rs.ProductKey,rs.OrderDateKey,rs.SalesOrderNumber;
OVER Clause Windowing • Some existing queries do not optimize well • Example: Details of orders and days since previous order of each product -- Windowed approach SELECT ProductKey,OrderDateKey,SalesOrderNumber, OrderDateKey-LAG(OrderDateKey) OVER (PARTITION BY ProductKey ORDER BY OrderDateKey,SalesOrderNumber) AS DaysSincePrevOrder FROM dbo.FactResellerSales AS rs ORDER BY ProductKey,OrderDateKey,SalesOrderNumber;
OVER Clause Windowing • Some existing queries do not optimize well • Example: Details of orders and days since previous order of each product
Sequences • User-defined object • Not tied to any particular table • Not restricted to being used in a single table • Eases migration from other database engines • Does not support Transaction
Metadata Discovery • Applications need to determine metadata associated with code batches or stored procedures • New functionality provided as both system stored procedures and via DMVs − DMVs make the metadata easy to consume in a relational query − Additional DMV option for passing an object ID (helpful for stored procedures or triggers) System Stored Procedure Equivalent DMV sp_describe_first_result_set sys.dm_exec_describe_first_result_set sys.dm_exec_describe_first_result_set_for_object • EXECUTE supports a WITH RESULT SETS clause
Enhanced Function Library • New functions added Conversion Date and Time Logical and String Analytic PARSE DATEFROMPARTS CHOOSE CUME_DIST TRY_PARSE TIMEFROMPARTS IIF PERCENTILE_DIST TRY_CONVERT DATETIMEFROMPARTS CONCAT PERCENTILE_CONT DATETIME2FROMPARTS FORMAT PERCENT_RANK SMALLDATETIMEFROMPARTS FIRST_VALUE DATETIMEOFFSETFROMPARTS LAST_VALUE EOMONTH LEAD LAG • LOG function now supports an optional base
General T-SQL-Related Enhancements • THROW statement − Can throw user-errors − Can rethrow any errors in a CATCH block (including system errors) • Extended Events supports many new events − Mostly relating to memory page allocation
WWW.KOLKATAGEEKS.COM WWW.SQLSERVERGEEKS.COM

More Related Content

SQL Server2012 Enhancements

  • 1. By Abhishek Sur Microsoft MVP, Client App Dev Follow us at Twitter @abhi2434 & @sqlservergeeks
  • 2. Why SQL Server 2012 ?  Always On  Contained Databases  Column Store Indexes  Visual Studio integration with Management Studio  TSQL Enhancements 2
  • 3. 3
  • 4. The Concept RowStore Page Int Varchar Datetime float • ColumnStore reduces redundancy of data and compression as the data type remains same for the entire page. • All data for a single column could be retrieved from a non clustered Columnstore Page index page. Int Int Int Int float float float float Limitation • 1 Table can have one columnstore index. 4
  • 5. 5
  • 6. Introducing SQL Server 2012 Transact-SQL Improvements • Query Pagination • OVER Clause Windowing • Sequences • Metadata Discovery • Enhanced Function Library • General T-SQL-Related Enhancements
  • 7. Query Pagination • Common application requirement to retrieve rows − For a given page − For a given page length − Our Options ??? − Retrieve all and filter in application end − Retrieve data using Row_NUMBER function − Any other options ? − Denali gives better option for pagination
  • 8. OVER Clause Windowing • Some existing queries do not optimize well • Example: Details of orders and days since previous order of each product -- Traditional approach SELECT rs.ProductKey,rs.OrderDateKey,rs.SalesOrderNumber, rs.OrderDateKey- (SELECT TOP 1 prev.OrderDateKey FROM dbo.FactResellerSales AS prev WHERE rs.ProductKey=prev.ProductKey AND prev.OrderDateKey<=rs.OrderDateKey AND prev.SalesOrderNumber<rs.SalesOrderNumber ORDER BY prev.OrderDateKey DESC, prev.SalesOrderNumber DESC) AS DaysSincePrevOrder FROM dbo.FactResellerSales AS rs ORDER BY rs.ProductKey,rs.OrderDateKey,rs.SalesOrderNumber;
  • 9. OVER Clause Windowing • Some existing queries do not optimize well • Example: Details of orders and days since previous order of each product -- Windowed approach SELECT ProductKey,OrderDateKey,SalesOrderNumber, OrderDateKey-LAG(OrderDateKey) OVER (PARTITION BY ProductKey ORDER BY OrderDateKey,SalesOrderNumber) AS DaysSincePrevOrder FROM dbo.FactResellerSales AS rs ORDER BY ProductKey,OrderDateKey,SalesOrderNumber;
  • 10. OVER Clause Windowing • Some existing queries do not optimize well • Example: Details of orders and days since previous order of each product
  • 11. Sequences • User-defined object • Not tied to any particular table • Not restricted to being used in a single table • Eases migration from other database engines • Does not support Transaction
  • 12. Metadata Discovery • Applications need to determine metadata associated with code batches or stored procedures • New functionality provided as both system stored procedures and via DMVs − DMVs make the metadata easy to consume in a relational query − Additional DMV option for passing an object ID (helpful for stored procedures or triggers) System Stored Procedure Equivalent DMV sp_describe_first_result_set sys.dm_exec_describe_first_result_set sys.dm_exec_describe_first_result_set_for_object • EXECUTE supports a WITH RESULT SETS clause
  • 13. Enhanced Function Library • New functions added Conversion Date and Time Logical and String Analytic PARSE DATEFROMPARTS CHOOSE CUME_DIST TRY_PARSE TIMEFROMPARTS IIF PERCENTILE_DIST TRY_CONVERT DATETIMEFROMPARTS CONCAT PERCENTILE_CONT DATETIME2FROMPARTS FORMAT PERCENT_RANK SMALLDATETIMEFROMPARTS FIRST_VALUE DATETIMEOFFSETFROMPARTS LAST_VALUE EOMONTH LEAD LAG • LOG function now supports an optional base
  • 14. General T-SQL-Related Enhancements • THROW statement − Can throw user-errors − Can rethrow any errors in a CATCH block (including system errors) • Extended Events supports many new events − Mostly relating to memory page allocation