Please start any new threads on our new 
    site at https://forums.sqlteam.com.  We've got lots of great SQL Server
    experts to answer whatever question you can come up with.
    
        | 
                
                    | 
                            
                                | Author | Topic |  
                                    | albertkohlAged Yak Warrior
 
 
                                        740 Posts | 
                                            
                                            |  Posted - 2013-07-23 : 10:48:19 
 |  
                                            | i know this is kinda vague, but hopefully someone can help me. i have a SQL Server that has 32 cores (64 logical). and for some reason, when we have a job that kicks off that uses 39 of those 64 CPU's and basically tacks them out. when that happens the SQL Service chokes for about 5 minutes until it's done doing what it needs to do. any ideas? |  |  
                                    | James KMaster Smack Fu Yak Hacker
 
 
                                    3873 Posts | 
                                        
                                          |  Posted - 2013-07-23 : 11:03:13 
 |  
                                          | That when you have a job that takes up a large number of CPU's, things slow down seems to be a natural cause/effect. One thing you might consider is reducing the parallelism by setting MAXDOP to a lower number.  Sometimes when you have too high a degree of parallelism, it can work to your disadvantage. http://msdn.microsoft.com/en-us/library/ms181007(v=sql.105).aspx |  
                                          |  |  |  
                                    | albertkohlAged Yak Warrior
 
 
                                    740 Posts | 
                                        
                                          |  Posted - 2013-07-23 : 11:13:54 
 |  
                                          | thanks. i'll take a stab @ that. |  
                                          |  |  |  
                                |  |  |  |