Author: Kevin Feasel

The Recording

The Panelists

• Kevin Feasel
• Mala Mahadevan

Notes: Questions and Topics

Google Cloud Outage Leads to Internet Outage

The first major topic was a quick reminder of how tightly woven cloud services are to our lives, even when we don’t use them. Google Cloud had a service outage of their key-value service. This affected anybody who used those services, including Cloudflare. Cloudflare’s own services were up, but because they had a dependency on Google’s KV service, it didn’t matter, and to us, 20% of the world’s Internet traffic went down for several hours. Not a great day for a lot of people.

More Tests of ZStandard Backup Compression in SQL Server 2025

The second topic we covered was an Aaron Bertrand article on the upcoming ZStandard backup compression format in SQL Server 2025. We looked at Aaron’s numbers and compared them to what we had seen from Anthony Nocentino and Andy Yun earlier. So far, there’s no strong concurrence in what’s “best.” Even where Andy and Anthony found that setting ZSTD compression to High was uniformly bad, Aaron wasn’t able to replicate that in his circumstances. I look forward to additional testing, including from people whose first name does not begin with the letter A.

Microsoft Build Leaving Seattle

To wrap up the day, Mala brought up that Microsoft Build will not be hosted in Seattle in the year 2026. Early reports were that San Francisco might be an alternative destination, though based on what we learned after the show ended, the authors of those articles edited out that information. Build has been in Seattle since 2017 and it was in San Francisco the few years prior.

The Recording

The Panelists

• Kevin Feasel
• Mala Mahadevan
• Mike Chrestensen

Notes: Questions and Topics

Microsoft Build Announcements

The first major topic was a quick review of Microsoft Build announcements with respect to the data platform. We started off with James Serra’s blog post on the topic and then spent some time in the Microsoft Build Book of News. The nicest thing we said is that, well, hey, one of the topics wasn’t around Copilot or Fabric.

SQL Server Management Studio 21 Now Available

Mike wanted to let us know that SQL Server Management Studio 21 is now available. Mike talked a little bit about some functionality changes and we spent a bit of time discussing SSMS switching over to the 64-bit Visual Studio shell and how Visual Studio and Visual Studio Code are very different beasts.

Backup Compression in SQL Server 2025

After that, I covered a pair of blog posts on the new ZSTD compression format for backups in SQL Server 2025. Anthony Nocentino and Andy Yun had blog posts on the topic that we covered. I still want to caution against taking their numbers as the sum total of experience here, as there are a lot of factors that go into database compression ratios and time to process. But it’s great that these gentlemen have given us some data points that we can use as a starting point.

Describing Result Sets

Finally, we made Mike’s week by teaching him about the sp_describe_first_result_set stored procedure and the related sys.dm_describe_first_result_set() function. Mike needed a way to get column names and data types from an arbitrary T-SQL query, and this has been in the product since SQL Server 2012, so there’s (almost) no version concerns.

The Recording

The Panelists

• Kevin Feasel
• Mala Mahadevan
• Mike Chrestensen

Notes: Questions and Topics

SQL Saturday Raleigh Recap

The first major topic was our SQL Saturday recap. I spent a bit of time talking about the event, as well as trying to provide some advice for people interested in running their own events. I didn’t have anything scripted out, so it probably came out as a jumbled mess, but them’s the breaks on Shop Talk.

Stack Overflow and Its Discontents

Our other major topic was around the decline in fortunes for Stack Overflow. We talked a bit about it, and what has likely replaced it: a combination of LLMs and Discord servers. That’s not a long-term stable solution (and I don’t think Reddit is the answer, either), but I mentioned the challenges of keeping a forum-style website going over a long time: you have people who form part of the institution, and as they age out or leave, it’s hard to keep things going. I expect that, over the long term, we’ll have something similar again. And admittedly, Stack Overflow isn’t dead, though it is on the ropes.

But hey, at least the Stack Exchange people know what will fix their problems: rebranding.

The Recording

The Panelists

• Kevin Feasel
• Mala Mahadevan
• Mike Chrestensen

Notes: Questions and Topics

SQL Saturday Raleigh

Our big news is around SQL Saturday Raleigh, which we’ve officially launched. It will take place on May 17th at Duke Health in Durham. You can register at the SQL Saturday website. There are free tickets that do not include lunch, and paid tickets that include lunch. We will not have lunch available for sale on the day of the event.

We’re also hosting Brent Ozar for a pre-con the day before. You can sign up for this on Brent’s website.

Azure ML SDK/CLI v1 Deprecation Notice

Microsoft has announced that the Azure ML Python SDK v1 will officially go away on June 30th of 2026, and the az cli AML extension v1 will go away in September of 2025. We talk a little bit about what that means, and the upgrade guide that Microsoft has provided. I also teased a video that will drop next week, on April 29th.

Iceberg vs Delta Lake

Our other major topic was a comparison of Apache Iceberg vs Delta Lake. This was a topic Mike wanted to know more about, and I happened to read (and curate) a great comparison of the two products from Maria Zakourdaev, whose name I clearly cannot pronounce. Maria does a great job of laying out the pros and cons of each platform and I highly recommend checking out that article if you’re interested in Iceberg.

The Recording

The Panelists

• Kevin Feasel
• Mala Mahadevan
• Mike Chrestensen

Notes: Questions and Topics

SQL Saturday Raleigh

Our big news is around SQL Saturday Raleigh, which we’ve officially launched. It will take place on May 17th at Duke Health in Durham. You can register at the SQL Saturday website. There are free tickets that do not include lunch, and paid tickets that include lunch. We will not have lunch available for sale on the day of the event.

We’re also hosting Brent Ozar for a pre-con the day before. You can sign up for this on Brent’s website.

The State of Microsoft Fabric

I used a very lengthy Patrick LeBlanc blog post covering Microsoft Fabric updates as a springboard for quite a few pent-up thoughts around Microsoft Fabric. I want to thread that needle of “This is going to be really cool two years from now” without pretending that the product is more than it currently is.

Oracle Admits to Data Breach

The final article comes from Mala: Oracle has admitted to a data breach in their cloud Gen1 platform.

The Recording

The Panelists

• Kevin Feasel
• Mike Chrestensen

Notes: Questions and Topics

SQL Saturday Raleigh

Our big news is around SQL Saturday Raleigh, which we’ve officially launched. It will take place on May 17th at Duke Health in Durham. You can register at the SQL Saturday website. There are free tickets that do not include lunch, and paid tickets that include lunch. We will not have lunch available for sale on the day of the event.

We’re also hosting Brent Ozar for a pre-con the day before. You can sign up for this on Brent’s website.

ETL, ELT, and PowerShell

Our primary question for the day came in via e-mail from a viewer:

I currently use PowerShell scripts to pull data from source APIs for my company’s data pipelines. Our BI department is nascent, thus our scripts are mostly just pulling data and loading it into a tabular format in SQL Server. We don’t do much in the way of transformations. This has caused issues with large memory grants and implicit conversions in our SQL Server.
I’m curious if PowerShell is an appropriate or even adequate tool for performing transformations? If not, what other options are there?
We are trying to keep as much “processing” as possible outside of SQL Server to reduce licensing costs.

Mike and I spend a bit of time talking about viable options, areas of focus, and industry trends.

The QUALIFY() Clause and Its T-SQL Equivalent

Mike wanted to know about the QUALIFY() clause that exists in Teradata and Snowflake. I talk a bit about its purpose and reference a Stack Overflow post that shows some T-SQL equivalents. QUALIFY() looks pretty neat and I certainly wouldn’t mind it being available to us in T-SQL, but I won’t hold my breath on it, as there are functional alternatives.

A Minor Rant about Oracle Date Formatting

Our other topic of the night was a rant from Mike about how Oracle’s date formatting functions are annoying. I have a tiny amount to add, but this was mostly Mike’s time to shine.

The Recording

The Panelists

• Kevin Feasel
• Mala Mahadevan
• Mike Chrestensen

Notes: Questions and Topics

SQL Saturday Raleigh

Our big news is around SQL Saturday Raleigh, which we’ve officially launched. It will take place on May 17th at Duke Health in Durham. You can register at the SQL Saturday website. There are free tickets that do not include lunch, and paid tickets that include lunch. We will not have lunch available for sale on the day of the event.

We’re also hosting Brent Ozar for a pre-con the day before. You can sign up for this on Brent’s website.

Handling Large Data Exports

Mike brought up our next topic: what are some tools and techniques for handling large-scale data export from databases for data scientists? We talked a bit about classic ETL tools like Informatica, SQL Server Integration Services, and Ab Initio. I also talked a bit about file formats, dealing with Parquet files as a standard whenever possible, and more.

Red Gate 2025 State of the Database Landscape

The final topic had us look at the highlights of Red Gate’s 2025 State of the Database Landscape. The full report is available but I stuck to the notes on Red Gate’s website. If you want the full report, go check it out.

The Recording

The Panelists

• Kevin Feasel
• Mala Mahadevan
• Mike Chrestensen

Notes: Questions and Topics

Notes on NOLOCK

Mala sent in the first link, an article from Joe Fleming regarding NOLOCK. In it, Joe makes the argument that ceteris paribus, NOLOCK is not a good thing, but that there’s little to no value in eliminating it in an environment. I agree with the former and strongly disagree with the latter. Along the way, we talk for a bit about Read Committed Snapshot Isolation as the solution in most circumstances.

2025 Data Professional Salary Survey Results

The other point of discussion was Brent Ozar’s 2025 Data Professional Salary Survey results. I popped open Excel and did a little bit of pivot table work. We talked a bit about data cleanliness, salary extremes, trends, and why mean can be misleading for things like salary.

The Recording

The Panelists

• Kevin Feasel
• Mala Mahadevan
• Mike Chrestensen

Notes: Questions and Topics

TriPASS Elections

Our TriPASS elections have wrapped up and our intrepid board members have survived for another two-year term.

Mala’s Thoughts on PASS Summit 2024

Mala then talked a bit about her experiences from PASS Summit 2024. She’s been traveling for a bit and this is her first episode back, so we had a chance to catch up with her a bit.

Thoughts on SQL Server 2025

We spent a lot of time this episode on SQL Server 2025, talking about some of the new functionality based on what we heard at Ignite in 2024 and thoughts for how this affects data engineers. Along the way, I generated a cocktail napkin level drawing of how vectorization works and we spent quite a bit of time talking about the new JSON data type and how this is a bit of a reversal from “the XML data type was a mistake and we should just keep JSON in NVARCHAR” thoughts in the past.

The Recording

The Panelists

• Kevin Feasel
• Mike Chrestensen

Notes: Questions and Topics

TriPASS Elections

It’s election season for TriPASS and we have a slate of electors up for vote. If you are a TriPASS member, you’ll have received an e-mail about this.

The Power of Anti-Semi-Joins

I learned something from Erik Darling. I’ve learned many things from Erik Darling, but this one was around NOT EXISTS versus 0 = (SELECT COUNT(*)...). It turns out that SQL Server has an optimization that makes both return the same execution plan. After the recording, I went and checked Postgres, which does not have the same optimization. This is yet another case of how behaviors in one platform may subtly differ from others.

Approximate Functions

Mike successfully side-tracked me and we talked a bit about the three approximation functions in SQL Server: APPROX_COUNT_DISTINCT, APPROX_PERCENTILE_CONT, and APPROX_PERCENTILE_DISC. I explained a bit about how they work, linked to some of the original academic papers, and described the difference between the continuous and discrete functions.

Tenant Switching in Microsoft Fabric

Our final topic of the night was from a Koen Verbeeck blog post, covering tenant switching in Microsoft Fabric and the Power BI Service. This is another example of the classic Microsoft pattern of developing something without designing in the ability to switch tenants / subscriptions / accounts.