Let’s hang out in the backyard – as recently seen on Zillow Gone Wild – and take your top-voted questions from https://PollGab.com/room/brento.

Here’s what we covered:

• 01:09 AussieDBA: I was saddened to hear of the passing of Andrew Clarke, who surely had the best pseudonym in the SQL industry, Phil Factor. His articles were some of the first I read online, and I was impressed with his complete and clear explanations. Can you share your fondest memory of him?
• 03:06 Elijah: Parallelism waits are the most common wait type on our server. My Cost Threshold for Parallelism is set to 50 and MAXDOP is set to 8. I want to gradually lower MAXDOP to reduce waits. How can I convince our DBA who is very hesitant to tune server-wide settings and prefers using Resource Governor?
• 04:27 DMExecCoffeeStats: Back in February, you said something to the effect of “you shouldn’t use schemas to namespace tables.” If I understood you correctly, is there a reason to prod my coworkers into undoing this practice, or should I just leave it be?
• 06:04 WhoMovedMyCheese: We’re evaluating a couple of different monitoring systems. Do you have any preference or experience?
• 06:48 AubreyPlaza: My Azure SQL DBs are constantly having a nervous breakdown from constant INSERT/UPDATE spam. I think you mentioned Stack Overflow batching stuff instead of being chaos gremlins. Which course of yours teaches you how to do batching?
• 07:38 Elwood: Any tips for finding fast and reliable internet when you’re traveling, like overseas or cruise ships?
• 08:53 Daniel: In your post about max memory settings, you casually mention that you set this up automatically on your lab servers as part of Agent startup. Are there any other settings you use in this context, and would you recommend this in production environments as well?
• 10:23 Dopinder: Do you recommend any AI courses that are complimentary for a SQL DBA?
• 11:27 NotCloseEnoughToRetirementToStopLearning: Has setting a persisted sampling rate for a statistic ever gotten you across the finish line? If so, please share the problem that it solved and how you determined the right sampling rate.
• 12:18 MyTeaGotCold: Have you played around with SQL Server 2025’s new Optimized Locking yet? Any thoughts?
• 15:08 DrawingABlank: What happens when a memory-optimized table runs out of memory?
• 15:53 It Could Be a Boat: If you were going to start your career all over again, would you pursue expertise in Postgres as opposed to SQL Server?
• 16:47 FrugalShaun: I’ve just finished my first Azure SQL DB consulting job. Tools like sp_Blitzcache were useful but I found the experience painful. Simple things like tracing queries was much more difficult than it should have been and the docs suck! How do you cope or do you just avoid?
• 17:53 Benjamin: I just started a new job as Jr. DBA. I was shocked to discover on my second day that there are several stored procedures on their SQL Server that use linked servers to import data from oracle databases that they don’t own. What alternatives can/should I offer to leadership?
• 19:09 Dopinder: Does high VLF count (4,160) ever matter for TempDB? It’s on azure vm ephemeral drive.
• 20:09 T-Rex Come Back: Ever seen a contained database out in the wild? Not the AG thing. Seems like a dead feature.
• 20:35 UpdateStatsFixesAll: Hi Brent, I have a query plan which is showing a “Columns With No Statistics” warning. I can see a statistic for this column below that table in Object Explorer and when I update that statistic the warning in the plan disappears. Any idea why the statistic is being ignored?
• 21:31 Big Blue Couch: Hey Brent, Do you have any thoughts on MS use of CDC to mirror on prem DBs to Fabric in SQL Server versions prior to 2025?
• 23:05 It Could Be a Boat: Hey Brent, do you have any advice on working for a micro-manager or control freak?
• 25:01 DBA_Unsupported: My new employer is small shop with a lot of ”tech debt”. Mostly SQL2008 & 2012 instances. They won’t upgrade anything because the environment is stable. Any recommendations for the smoothest sailing?
• 27:13 Dopinder: Is asking programming questions to AI just as helpful as reading the documentation?