SQL Server Query Plan Analysis: The 5 Culprits That Cause 95% of Your Performance Headache

Ever open a huge query plan to track down a performance issue and instead find yourself overwhelmed by the sheer number of objects on the screen? How do you figure out which part of the plan.

[drstrangebudgie]

0:00:00 intro
0:05:30 how query plans work
0:17:59 top five quick check iterators
0:26:47 lookup
0:34:51 spool
0:44:18 sort
0:51:13 hash match
0:55:45 serial nested loops
1:02:41 scan (seek predicates vs residual predicates)
1:11:28 outro

[BalvinderSingh-uhmy]

Best video on performance without question, I was really worried I thought I'd save this video. It was out before this date. Anyway super happy that I found it again, I'm amazed and sadly shocked that I’m the first person to leave a thanks it's truly excellent. Forget Moore’s law if every DBA saw this and understood it massive impact on the world speaking like the nerd that I am.

[Rafian]

This is the ultimate performance tuning video. Hats off to you!!

[alexanderfleming]

I cannot say enough great things about this video. Great job

[massefromla]

I learned a lot from this video, thanks .. big thumbs up 👍

[mokarem]

Many thanks for uploading such an informative lecture.

[lazykitten]

Can anyone explain to me why he keeps using "cross apply" instead of 'inner join'? there are no tablar functions.

[monday]

02:00 That is why, in 2035, the database will be tuning itself : the fixing mechanism is always the same, and is always logical. It could have been fixed by now, but programmers focus now lies on designing nice looking apps for phones that work like s

[jez]

This why I'm not a DBA. It takes me a long time to understand why changing a SQL query results in the change of the execution plan that it does, and this guy skips over things way too quickly for me to really grasp.

[codewiserelizondo]

Many thanks for sharing! is the sql shown available somewhere?

[Maxvm]

Can someone explaing me more why we shouldn't take into consideration %cost ? On EDX there is a course about query optimization and one of the first things during the query execution plans introduction is to look at %cost. Where is the catch :D

[evald]

@30min is shown how a query that took 1sec, after added an empty column, took 17 sec. How can be possible? this is not clear at all in the video or maybe i missed something