Optimizing tempdb configuration with SQL Server 2012 Extended Events
One of the most obvious bottlenecks in the performance of tempdb is caused by PAGELATCH, in-memory latch contention on the allocation bitmap of each data file used. We can use one of the rules-of-thumb to choose what should be roughly the best number of files, but how then do you check to see whether you've got it right for your data and workload?
