Performance Implications of Parallelism

Wednesday, December 5th, 2012 12:00 PM Central Time


Parallelism is a key factor in the performance of large queries. It allows SQL Server to perform a large operation quickly and yet at other times can appear to hinder performance. It is a complex topic that can often be difficult to troubleshoot and tune. This session will explain the how parallelism works, why it is good, and when it is bad. Some of the key takeaways will be:

  • What are CXPacket waits and should we care about them?
  • How parallel tasks are scheduled
  • How to read a parallel execution plan
  • How developers can tune for parallelism
  • How DBAs can tune for parallelism

 

Event Registration

Rate and Comment
  Select Registration Options
 Select the items you want below.
Performance Implications of Parallelism
EVENT is Closed
Rate and Comment
Rate and Comment
  Please Confirm Your Information
Email:
 
 
Desired/Current Password:
 
New members continue filling out the form.
Returning Members Login now!
First Name:   Last Name:   Phone:    
Address Line 1:   Address Line 2: Postal Code:  
City:   State:   Country:  
Company:   Title/Position:  
Rate and Comment
Updating Registration
Submitting Registration


Sponsors

Speakers

  • Robert L.  Davis

    Robert is a Sr. Product Consultant and Chief SQL Server Evangelist for Idera Software. Previously, he was the Program Manager for the SQL Server Certified Master Program in Microsoft Learning. He was also a Sr. Production DBA at Microsoft with more than 12 years’ experience with SQL Server -- author of Pro SQL Server 2008 Mirroring -- writer for SQL Server Magazine -- Microsoft Certified Master: SQL Server 2008 -- Speaker/trainer.

About Registration

  • Your registration includes:
    • On-demand viewing for 30 days following the event
    • Access to the virtual event video content
    • Live, interactive chat with other attendees, speakers and sponsors
    Not sure you can attend? No problem, all the video content will be available on-demand for 30 days.