Y’all posted & upvoted questions at https://pollgab.com/room/brento, and we finished up the session talking about the relationships between developers and database administrators.

Here’s what we covered:

• 00:00 Start
• 02:35 Nortzi: Hi Brent. Is there a way to optimize sorting a result using a column from a different table? Using POC index pattern (partition,order,covering) works great to eliminate sort operations on columns from the same table but what if I have a column from another table in the order by?
• 04:25 Gustav: Who is winning the serverless cloud SQL race? In addition to price, what are the big differentiators?
• 05:38 Pool_Party: You wrote on your blog that SHRINKDATABASE is not good because creates fragmentation and therefore the problem is not solved. But I’m in an Azure SQL Elastic Pool scenario and I need to claim more space otherwise databases will not fit. What to do?
• 07:29 SD: Hi Brent, I have Non-Prod AGs that I want to upgrade from SQL 2016\2017 to SQL 2022 (And downgrade the Edition from Enterprise to Developer). Can I do this by adding SQL 2022 Developer Edition Nodes to the AG and doing a failover (is this supported \ allowed \ known issues)?
• 08:56 Deepak: Do you think generative AI would be a good addition to Redgate SQL Prompt (i.e. help write queries)?
• 11:13 Jahid Ajmeri: Hi Brent, I am DBA. I want to use transactional replication with updatable subscription. There are multiple services insert/update data every 5 second into database. I want to make this write operation on both pub & sub database without conflicts. What you suggest.
• 14:01 Adam: Hi Brent, I’m finding on occasional that my msdb is going into Suspect mode around the same time in the evening. I have run DBCC and VM disk checks and nothing has been found. Have you ever come across this before and if so, how did you solve it?
• 15:27 Dwayne: Which games do you enjoy on the PC? Which games do you enjoy on the PS5?
• 16:56 TomInYorks: I’m exploring solutions to locking/blocking seen in an application (multi-tenant, DB per customer but customisable reporting on live data) – was looking at RCSI or Snapshot Isolation. Can you recommend a good place to start when analysing the suitability of either of those?
• 18:31 Vishnu: Is Apple Worldwide Developers Conference or Microsoft Bing the more influential tech conference?
• 18:57 Vineeth: What is your opinion of the new Json native data type in Azure SQL DB?
• 20:06 Deepak: Automated weekly Qualys cyber scans attempts to penetrate the SQL Server and reports findings back to corporate. One of the test threads attempts to connect to the DAC. Do you see any risks in this?
• 20:57 Edwardo: Is there such thing as a “memory deadlock”? I recently observed multiple parent queries snagging 25% memory apiece and then calling a linked server loopback connection to spawn some child queries under a different SPID, and the child queries were waiting on RESOURCE_SEMAPHORE.
• 22:21 Dom: Hi Brent, You mentionned a couple of time about “san backup” for SQL large databases. I google it and didn’t find much information about how that “work”. I’m guessing it’s backing up the mdf and ldf (it then cannot take log backup right?) Does it mean DB needs to be in simple ?
• 23:04 Halldora: For sp_BlitzCache @MinutesBack argument, is there good way to tell how many minutes back we can go before data is no longer present?
• 23:58 Yitzhar: Do you have any good use cases for the SSMS option : “Discard results after execution”?
• 24:41 toepoke: Hey, I use lookup tables in my db, typically with “Id, Code, …” columns, with a FK onto “Id” from the parent table. “Code” should be unique. “StatusCode” has a unique nonclustered index to avoid dupes. Is this more icky than dropping “Id” & using “Code” as the identity col?
• 25:31 Dwayne: What MSExcel skills (if any) do you recommend having for the SQL DBA?
• 27:14 dave: Some developers say that DBA stands for “Don’t Be Asking” – How do you feel about Dev-DBA relationships?