Five stages of grief – internals of a hash spill
This page contains the description for my conference session “Five stages of grief – internals of a hash spill”, and links to the slide deck used in this presentation.
|Target audience||Experienced database developers and DBAs, plus all curious geeks.|
|Short description||A deep dive into the internals of the various stages that occur when a Hash Match operator spills.|
|Duration||The ideal length for this session is 60 minutes.|
|Full abstract||You know the Hash Match operator. You know it requires a lot of memory. And you also know that it sometimes needs more memory than it has, in which case it spills to tempdb and is slow.
But do you know EXACTLY what happens under the cover in the case of a hash spill? Or does you knowledge stop at “it gets slow”?
Unless you work for Microsoft on the SQL Server engineering team, there really is no need to know the internals of a hash spill. But that doesn’t mean you can’t be curious! Would you like to waste some time to learn about dynamic role reversal, grace hash join, bail-out, bit-vector filtering, and more? Then join this session!
Click here to download a ZIP file containing the slide deck and demo code used in this presentation.
(Note: my presentations tend to evolve over time, so the version you find here may not be an exact match for the version you witnessed).