SQL Server Query Processing Internals

When faced with performance issues there are many lines of attack.  Tuning the engine itself can get you so far, however for maximum effect you need to understand how the engine translates SQL statements into performable actions.  This is not a simple task, it is a massive task to deal with a multi-table join and the number of permutations can be immense. 


In this precon, we will be looking at the processes involved of taking a simple statement and forming a workable well performing execution plan.


Firstly, we will be looking at Execution plans and picking them apart to describe the actions that are executed by the engine to perform the actions required by the query.  This will be followed by statistics, estimation and caching and the role that they play in forming and costing execution plans.


After lunch,  we will see how SQL Server 2014 has new functionality within it and the problems that they are designed to resolve.  Finally, they day will finish with query optimizer internals, picking apart its various internal steps and forensically examining its operations.
  • Downloads
    Sorry, there are no downloads available for this session.
  • SpeakerBIO
    dave_ballantyne.jpg

    Dave is a freelance SQL Server database architect / designer / developer / consultant and lives near London, England.  He is a regular speaker at UK and European events and user groups and currently is supporting the London SQL community by organizing the "SQL Supper" user group.  He takes a keen interest in all things SQL and data related and is never happier than picking apart a poorly executing query.  He has also created an open sourced add-in for SQL Server data tools (SSDT) called "TSQL Smells".  This detects and reports on 'suspect'  code within a project and can be found at http://tsqlsmellsssdt.codeplex.com . Outside of work he is husband, father of 3 and exasperated archer.

    http://www.clearskysql.co.uk http://dataidol.com/davebally/feed/
  • Video
    The video is not available to view online.