Press "Enter" to skip to content

Author: Kevin Feasel

Last Page Insert Contention in SQL Server

Published 2023-07-05 by Kevin Feasel

Eitan Blumin spots a wild Latch Convoy:

The “Last Page Insert Contention” in SQL Server, also known as “Latch Convoy Problem“, also known as “PageLatchEx Contention” is one of those extremely rare use cases that are very difficult to see in real-world scenarios.

Evidently, it was impactful enough that Microsoft implemented a solution for this problem back in SQL Server 2019 in the form of the new OPTIMIZE_FOR_SEQUENTIAL_KEY index option, which reportedly fixes it.

Click through to learn more about a scenario in which Eitan saw this in the wild. In fairness, I’m not sure I’d do any better at realizing that this was a last page insert contention problem.

Comments closed

A Primer on Databricks Unity Catalog

Published 2023-07-03 by Kevin Feasel

Beginner’s Hadoop gives us an overview:

The Databricks Unity Catalog is a feature provided by Databricks Unified Data Analytics Platform that allows you to organize and manage metadata about your data assets, such as tables, databases, and views. It provides a centralized metadata repository that enables users to discover, understand, and collaborate on data assets within a Databricks environment. The Unity Catalog integrates with various data sources and supports different metadata management capabilities.

Read on for an overview of what it does.

Comments closed

Listing Available Properties in Azure Data Factory

Published 2023-07-03 by Kevin Feasel

Andy Leonard builds a list:

Did you know Azure Data Factory (ADF) will actually list available properties? It will. One of the things I cover in my ADF training titled Master the Fundamentals of Azure Data Factory is this handy troubleshooting tip.

Read on to see how, though I’d personally like something which is a bit faster than waiting for the thing to execute and getting back what my choices are.

Comments closed

On-Demand Loading and Direct Lake in Power BI

Published 2023-07-03 by Kevin Feasel

Chris Webb gives us the beginnings of an origin story:

For any Power BI person, Direct Lake mode is the killer feature of Fabric. Import mode report performance (or near enough) direct on data from the lake, with none of the waiting around for data to refresh! It seems too good to be true. How can it be possible?

The full answer, going into detail about how data from Delta tables is transcoded to Power BI’s in-memory format, is too long for one blog post. But in part it is possible through something that existed before Fabric but which didn’t gain much attention: on-demand loading. 

Click through for another blog post on the topic and an idea of how these tie together.

Comments closed

Loading Data from On-Premises SQL Server into Microsoft Fabric

Published 2023-07-03 by Kevin Feasel

Reitse Eskens spends an hour or so:

In my previous blogs, I’ve written about Fabric and all the cool things it can do. Thing is, my load tests were based on files. Either CSV or Delta. But in reality, a lot of data comes from an on-premises database server. In reality, you might connect to a SQL 2008 instance or maybe even older. Truth be told, I haven’t got an instance in that version/edition around anymore. So I had to use SQL Server 2019, a version I’m encountering more often nowadays.

For this blog, it won’t make much sense to create a humongous database and try to get all the data in. Fabric will cope, the major issue (in my experience) is the internet connection between my local database and the Fabric environment. One thing I’m really curious about is if Fabric will have the Link capability that was introduced for Synapse Analytics and SQL Server 2022.

There’s no Link capability currently available, so Reitse does the next-best thing and uses Fabric pipelines.

Comments closed

An Overview of Data Modeling

Published 2023-07-03 by Kevin Feasel

Nikola Ilic provides an overview of data modeling:

In recent years, I’ve done dozens of training on various data platform topics, for all kinds of audiences. When teaching various data platform concepts and techniques, I find one of the concepts particularly intimidating for many business analysts, especially those who are just starting their journey. And, that is the concept of data modeling.

This is a good introduction and does a particularly good job of explaining why we have logical and physical data models. I have one medium-sized quibble with an otherwise-great article: 3rd Normal Form is nowhere near sufficient for a logical data model, and I’d make the strong case (in fact, I do make that case) that 5th Normal Form should be the standard and that 3NF is an anachronism which you should entirely replace with Boyce-Codd Normal Form.

Comments closed

Rolling Correlation in R

Published 2023-06-30 by Kevin Feasel

Steven Sanderson tries out a function:

In the world of data analysis, time-series data is a common sight. Whether it’s stock prices, weather patterns, or website traffic, understanding the relationship between variables over time is crucial. One valuable technique in this domain is calculating rolling correlation, which allows us to examine the evolving correlation between two variables as our data moves through time. In this blog post, we will delve into the rollapply function and its capabilities, exploring its applications through a series of practical examples. So, let’s get started!

Click through for an example of how it works.

Comments closed

Events in Apache Kafka

Published 2023-06-30 by Kevin Feasel

Lucia Cerchie explains the key metaphor in an event-driven system:

We can speak broadly, maybe even a little philosophically, about what events are. Events are “things that happen,” or sometimes, they are otherwise defined as representations of facts. All data is, in a way, a result of humans trying to grok events. At the same time, I honestly don’t find the definition helpful if we leave it at this level. Do we ever design apps around things that don’t happen? (Don’t think about that too hard.) 

Let’s get concrete: Events that might affect real-time data pipelines and applications, including things like Pinterest saves, USPS address changes, ship coordinate updates, and credit card transactions. 

Read on for more about events and how they drive the design of Kafka topics and the applications which use them.

Comments closed

Thoughts on Fabric Data Wrangler

Published 2023-06-30 by Kevin Feasel

Gilbert Quevauvilliers tries out a tool:

I was going through my twitter feed and I came across this tweet where they spoke about the Data Wrangler Calling all #Python users! Have you tried Data Wrangler in #MicrosoftFabric?

I thought I would give this a try and that was the idea for my blog post. I honestly had no idea that firstly was this possible, but second that it is so easy for data wrangler to do all the hard work for me

I am going to demonstrate 2 transformations in this blog post, the first will be changing the d_date from date to datetime and then using the columns from examples I am going to create a new column where it concatenates 2 columns delimited with a double pipe command.

Read on for Gilbert’s thoughts.

Comments closed

Protecting Kubernetes Services

Published 2023-06-30 by Kevin Feasel

Boemo Mmopelwa gives us an idea of Kubernetes service types and how to secure them:

A Kubernetes service is a logical abstraction that enables communication between different components in Kubernetes. Services provide a consistent way to access and communicate with the application’s underlying components, regardless of where those components are located.

In Kubernetes the default type is ClusterIP. Services abstract a group of pods with the same functions. Services expose pods and clusters. Services are crucial for connecting the backend and front-end of your applications.

This is different from your containerized applications that you can deploy on Kubernetes

Comments closed