Addressing large memory grant requests from optimized Nested Loops
'Optimized Nested Loops (or Batch Sort) is effectively an optimization aimed at minimizing I/O during a nested loop when the inner side table is large, regardless of it being parallelized or not.
The presence of this optimization in a given plan may not be very obvious when you look at an execution plan, given the sort itself is hidden, but you can see this by looking in the plan XML, and looking for the attribute Optimized, meaning the Nested Loop join may try to reorder the input rows to improve I/O performance. You can read more about optimized Nested Loops in Craig Freeman’s blog.'...
Trackbacks
Die Kommentarfunktion wurde vom Besitzer dieses Blogs in diesem Eintrag deaktiviert.
Kommentare
Ansicht der Kommentare: Linear | Verschachtelt