I start today’s Office Hours by playing with the very underwhelming Github Copilot in ADS, then take your questions from https://pollgab.com/room/brento.
Here’s what we covered:
- 00:00 Start
- 04:06 Github Copilot in Azure Data Studio
- 24:36 Byron: Seeing plan cache queries for a long since retired table. What is best way to track down the app querying this table?
- 25:57 Eduardo: What is your opinion of ApexSQL SQL Plan viewer? Feels like a competitor to SQL Sentry Plan explorer.
- 26:33 SQLrage: Are their any tips or tricks (that aren’t out there regarding patching or page lock in memory) to prevent Non Yielding Scheduler errors related to IO issues? Seeing it a lot in a high transaction system and MS comes back stating IO problem is the root cause. Thanks brotha!
- 27:23 GP Geek: I’ve a case where Index REBUILD runs every week, but the DB tables keep showing a fragmentation of over 98% every day.. what’s wrong ?
- 28:05 Nicolai: What is your opinion of the Microsoft MVP award?
- 32:37 Mr. SqlSeeks: When using a UNION operator to join the results of multiple queries, does the order of the queries affect how SQL Server gets the data?
1 Comment. Leave new
Regarding the Github Copilot:
I don’t really see the sense in it. Writing the requirement with the correct words is as much as writing the query itself, particulalry. if you are using a SSMS addin (as Redgate SQL Prompt) with snippets / makros.
It may help people who have absolute no idea about SQL, but even those can’t really use it, because you really need to know SQL and your database to confirm, that the code is correct or why it is not working, if GitHub adds some PostgreSQL stuff to your queries.
And to be honest: training the Copilot with the queries in StackOverflow etc. may not be the best idea. Of course there are good solutions, but a ton of mediocre / bad ones too (because usually the helping guys are either not THAT good and/or don’t know exactly the database on the askers side)