A Fine Slice Of SQL Server
Paul Randal explains what a log record is:
The simplest definition of a log record: it describes a single change to a database. A single operation in the database may cause multiple changes, but each change will usually have its own log record to describe it. An example of this is updating a column in a single row—it will do the following:
Read on to see what it will do, what it looks like, and what kinds of log records exist.
Comments closedLouis Davidson lays out the facts:
Which brings me to the point. There is a myth that goes around that you need to place filters in your SQL statements as early in the statement as possible. Most of this is due to the wild misunderstanding of how a query is executed (versus how your query is processed, which I covered last week.) The actual issue here is that the concept of filtering early is actually true, but certainly not in the way it has been taught.
SQL is a fourth-generation language and implementations approach it. With fourth-generation languages, the actual query you write is not the thing that runs, and there is an entire process to interpret what you wrote and execute operations that meet the intent of your query in the most efficient manner.
Now, this is where someone chimes in and gives all of the circumstances in which T-SQL (or pick your variant) fails to live up to its fourth-generation heritage, such as particularly complex queries, nested views with multiple joins, you using mechanisms that force a specific plan, etc. This is because real life is messy, as Louis shows in some of the examples.
So what’s the point of the first paragraph, then? Because I never miss an opportunity to talk about language generations.
Comments closedNikola Ilic digs into a feature:
I hear you, I hear you: Nikola, that’s what deployment rules in Fabric Deployment Pipelines are for, isn’t it? Well, partly. But there’s a Fabric item built specifically to put an end to this whole genre of pain, and it’s the variable library. This article is the long version: what it is, how it’s wired together under the hood, who can actually consume it, when you should reach for it, when you absolutely shouldn’t, how it compares to the other parameterization features in Fabric, and a real telco demo to make it all concrete.
Click through for a deep dive into how it works.
Comments closedErik Darling here with Darling Data, and in today’s video we’re going to continue on the Learn T-SQL voyage that we have started, and I’m going to talk today a little bit about indexed view matching, because SQL Server is, let’s just call it a mature, or an experienced database engine, and is quite capable, at least in Enterprise Edition, Standard Edition, you do not pay the Microsoft Friendship Tax, so you will be taxed performance-wise, but is quite capable of matching base queries to an indexed view where the syntax matches in some way between them. So, usually exactly between them, not in some way, usually pretty close to just about what you would ask for.
Click through for several tips and, as you experience the frustration of consistently trying to make best use of the view’s index, be glad you’re not trying to get filtered views to work.
Comments closedGilbert Quevauvilliers takes advantage of source control:
With the new Lakehouse Auto-Binding capability in Notebook Git integration, Fabric can now intelligently preserve and resolve the binding between your notebooks and their attached Lakehouses as you move them across workspaces. This makes true multi-environment development and CI/CD workflows in Fabric significantly smoother and more reliable.
I am going to show you how to do this in the blog post below.
That is pretty nice, and Gilbert has a demo of the process, showing that it’s not particularly onerous
Comments closedGarry Bargsley solves a problem:
Many routine administrative tasks that have traditionally been handled by SQL Agent still need to be performed:
- Scheduled stored procedures
- ETL processes
- Report generation
- Data cleanup
- Monitoring and alerting
- Business process automation
However, Azure SQL Database does not include SQL Agent.
Garry provides several solutions, and I would add to it third-party job scheduling solutions. Granted, that’s usually an extra expense (whether due to fees or supporting a roll-your-own solution), but it’s on the table. And some of them are better than what SQL Agent has to offer, even if I do like the fact that there’s an okay option built-in for DBAs.
Comments closedAndy Brownsword tries out a new feature:
Dealing with endless formats for SQL scripts is a special type of pain you get to enjoy as a data developer. With the arrival of SSMS v22.7 last week, we have a solution in the form of a T-SQL formatter which is now available in preview.
Read on for Andy’s thoughts.
Comments closedAbout ten years ago, I sketched out an idea for a different kind of SQL Server monitoring tool: one that gathered data just once per day, and gave you a short email with a specific list of actionable tasks to make a difference in health and performance.
Richie Rump did an amazing job of building SQL ConstantCare out in the years since, building a solution that was rock-solid and scaled well to terabytes of monitoring data. Every day, we sent thousands of emails for SQL Servers around the world.
This month, we shut off sales and began decommissioning it. Here’s why.
Click through for the reasons. Brent is still maintaining the First Responder Toolkit, so things like sp_Blitz will still be around. But I will miss the quarterly graphics of who’s using which versions of SQL Server, even if I always had to give the “This is a biased sample and may not be indicative of the entire population, but it’s still an interesting and informative sample” spiel each time.
1 CommentDejan Lukic tries to minimize personally identifiable information:
Large language models (LLMs) and the agents built on top of them ingest everything they are given, including personally-identifiable information (PII). In workflows where PII is inevitable, proper measures should exist for data sanitization.
Data can leak through model outputs, embeddings or even logs. Given that you have to use LLMs in your pipeline, in this article I will cover the anonymization techniques you can utilize in an LLM flow to minimize PII exposure vectors.
Some of the points are specific to language models (or at least storing data in vector databases), but other tips are more generic and can apply to classic data handling.
Comments closedReza Rad solves a common challenge:
Building a multilingual Power BI report has always been one of those requirements that sounded simple but turned out painful to implement. A French-speaking user opens your report and sees English category names. A Spanish-speaking user opens the same report and sees the same English values. You end up either maintaining multiple reports — one per language — or wrestling with external tools and complex workarounds just to get translations working properly.
Read on for an alternative solution that is—you guessed it—currently in preview.
Comments closed