Testing – Curated SQL

Dimensional Testing in Kafka

Published 2026-05-22 by Kevin Feasel

Most of my career in distributed systems has been as a tester, performance engineer and formal verification specialist. I’ve written performance benchmarking tools in the past, for RabbitMQ and Apache Pulsar but in recent years I’ve used OpenMessagingBenchmark (OMB) to run benchmarks against Apache Kafka and other messaging systems. But OMB is hard to deploy and has several limitations compared to more sophisticated benchmarking systems I’ve developed in the past. With Claude becoming so much better since Christmas I decided to write a Kafka-centric performance benchmarking tool, with a lot of inspiration from OMB. I took the bits I like about OMB and the things I like about the tooling I’ve built in the past, to make a performance testing tool for testing Apache Kafka.

Click through for an overview of the tool and how it works.

Migrating testthat to testit

Published 2026-05-18 by Kevin Feasel

Yihui Xie explains how to switch test frameworks in R:

Back in 2013, I wrote about testing R packages when I first released testit. Thirteen years later, I still believe that unit testing should be nothing more than “tell me if something unexpected happened.” Recently I converted a large testthat test suite to testit, and I thought I’d share a practical guide for anyone considering the same move.

Click through for that guide.

Storage Load Testing SQL Server

Published 2026-04-28 by Kevin Feasel

Anthony Nocentino designs a test:

I’ve been doing storage load tests for SQL Server for a long time, both as a consultant and now in my work at Everpure, and I see the same patterns over and over. Someone spins up a VM with two vCPUs, points it at a storage subsystem (cloud or on-prem), runs a thousand threads at it, and then concludes that the storage stinks. Or the opposite, where they buy a 64 gigabit HBA, plug it into the wrong PCIe slot, and wonder why they’re leaving half of the capacity on the table.

Designing a good load test isn’t about pushing the biggest number you can find. It’s about knowing what your gear is supposed to do, driving load until you find the point where the system breaks down, and measuring at the right places along the way. In this post, I’m going to walk you through how I think about designing a storage load test, the tools I reach for, and the pitfalls that will invalidate your results if you’re not paying attention.

Click through for Anthony’s thought process.

Comments closed

Testing Your SQL Server Backups

Published 2026-04-24 by Kevin Feasel

Vlad Drumea stresses the need to test:

Folks think that if they just take backups regularly, and their backup intervals match their recovery point objectives, then they can rest assured knowing (more like falsely thinking) they have backups from which to restore in case something bad happens to a database or to the entire instance.

Yeah, there are several things that can go wrong, even beyond the solid answers Vlad provides. Think about developers taking one-off backups that break the log chain, overly-aggressive purge processes that delete backups before they’re fully moved to long-term storage, etc.

Definitely check out Vlad’s post. If you’re interested, I also have a video series on YouTube that covers SQL Server backup and restore.

Comments closed

Unit Testing DAX via Semantic Link

Published 2026-04-13 by Kevin Feasel

Jens Vestergaard writes a test:

Every BI developer has felt it. You change a measure, update a relationship, or rename a column in a semantic model, and then you spend the next hour clicking through report pages to check if something broke. Manual spot-checking is how most teams validate DAX today. It works until it does not.

I have been building and maintaining semantic models for years. The further I get into Fabric-based development, the more my models start to feel like production code. They power dashboards that drive decisions. They feed downstream pipelines. When something breaks, the blast radius is real. And yet, the testing story has always been: deploy, open the report, squint at the numbers.

That gap bothered me enough to do something about it.

Click through to see what Jens has done.

Comments closed

Fixtures in Pytest

Published 2026-02-11 by Kevin Feasel

Jason Yousef shows off a capability in Pytest:

Pytest is one of those tools that feels obvious after you’ve used it for a bit. Tests are just functions. Assertions read like normal Python. And when you need context—database sessions, config, mock data—you reach for fixtures instead of duct tape.

Read on to see how they work. Admittedly, I don’t think I’ve used fixtures before in Pytest, but now seems like a good time to try it.

Comments closed

Coding against the Happy Path

Published 2026-02-11 by Kevin Feasel

Andy Brownsword thinks about successful code:

A common time to revisit old code is when something breaks. I was contemplating Pat’s question this week when a field length change had caused a truncation issue in an old SSIS data flow. Some code doesn’t age badly because it’s wrong, but because it didn’t expect to fail.

It’s all too easy to write a piece of code which ‘does the thing’, run a few variant tests, and send it on its way. But will it stand the test of time? That’s where my good and bad code diverge, and I usually revisit the code and find the old milk.

Admittedly, I’m not as good at this as I should be either. It can be a challenge to think through the possible issues that could arise and develop code to mitigate or eliminate those issues. But as Andy points out, it can be critical to success.

Comments closed

Workload Simulation in PostgreSQL

Published 2026-01-19 by Kevin Feasel

Dave Page announces a new load testing tool:

Most database benchmarking tools focus on raw throughput: how many queries per second can the database handle at maximum load? Whilst this is valuable information, it tells us little about how a system will cope with real-world usage patterns.

Consider a typical e-commerce platform. Traffic peaks during lunch breaks and evenings, drops off overnight, and behaves differently at weekends compared to weekdays. A stock trading application has intense activity during market hours and virtually none outside them. These temporal patterns matter enormously for capacity planning, replication testing, and failover validation.

Click through for more information, as well as a link to the pgEdge Load Generator GitHub repo.

Comments closed

Releasing a tSQLt Test Adapter for VS2022 and Later

Published 2026-01-05 by Kevin Feasel

Ed Elliott has been busy:

I have had a number of requests for me to update the tSqlt Test Adapter over the years so it would work with more recent versions of Visual Studio. I finally got around to doing this and I am pleased to say that the new version works with VS 2022, 2025 and should continue to work in future releases.

Between Visual Studio 2017 and 2019 the changes required were minimal but between 2017 and 2022 the changes meant a rewrite of the test adapter visual studio integration parts which meant it wasn’t a simple change.

Read on for information on how to install and use the extension.

Comments closed

Creating Test Data in Python via Faker

Published 2025-11-18 by Kevin Feasel

Brendan Tierney generates some artificial data:

A some point everyone needs some test data for their database. There area a number of ways of doing this, and in this post I’ll walk through using the Python library Faker to create some dummy test data (that kind of looks real) in my Oracle Database. I’ll have another post using the GenAI in-database feature available in the Oracle Autonomous Database. So keep an eye out for that.

Faker is one of the available libraries in Python for creating dummy/test data that kind of looks realistic.

Brendan generates some demo customer data, including an example of credit rating that allows for assignment of probability for each class.

Comments closed

Category: Testing