We upgraded our primary production SQL Server box a few weeks ago from a 2-processor/1-GB RAM/4 year-old server to a 4-processor/2-GB RAM/brand new server. We've consolidated a few more small database onto this new box, but it added less than 5% more transactions and connections.
Since the migration, the tempdb log file has been growing out of control...it reached 80-GB in less than 3 weeks. The total size of all the data files on the server is approx. 25-GB.
We've noticed 1 particular View, which utilizes a temporary table (@Temp_Table), will cause the tempdb log file to grow nearly 1-GB every time it's executed and it's been the only cause of real growth of the file since we started logging it's size (file size & space used) on Mon. When we execute this View on the old SQL Server box the tempdb log file doesn't always grow and when it does the space used eventually decreases too. In the 4+ years that we used the old box we never noticed the log file size and we had to address issues w/ other log files so we would have noticed if it were ever more than a couple gigs...the disks weren't that large.
We're confused on why the View is causing the tempdb log file grow so quickly and why so large, shouldn't the space used decrease after checkpoints?
Thx for the help!!!
Since the migration, the tempdb log file has been growing out of control...it reached 80-GB in less than 3 weeks. The total size of all the data files on the server is approx. 25-GB.
We've noticed 1 particular View, which utilizes a temporary table (@Temp_Table), will cause the tempdb log file to grow nearly 1-GB every time it's executed and it's been the only cause of real growth of the file since we started logging it's size (file size & space used) on Mon. When we execute this View on the old SQL Server box the tempdb log file doesn't always grow and when it does the space used eventually decreases too. In the 4+ years that we used the old box we never noticed the log file size and we had to address issues w/ other log files so we would have noticed if it were ever more than a couple gigs...the disks weren't that large.
We're confused on why the View is causing the tempdb log file grow so quickly and why so large, shouldn't the space used decrease after checkpoints?
Thx for the help!!!