Press "Enter" to skip to content

Day: July 28, 2026

Incremental Liquid Clustering in Apache Spark

Published 2026-07-28 by Kevin Feasel

Miles Cole lays out a recommendation:

Liquid Clustering was already a better abstraction than static partitioning due to its flexible nature. Fabric Spark Runtime 2.0 fixes the part that had me actively cautioning customers to reconsider blindly adopting it: the cost of maintaining the layout.

In Runtime 1.3 (Delta 3.2) a small append followed by OPTIMIZE would rewrite every file in a partial Z-Cube. In Runtime 2.0, the incremental strategy only touches files that are unclustered, small, or carrying a high density of deletion vectors. That changes Liquid Clustering from an occasional, potentially expensive maintenance operation into something that works beautifully with workloads of any shape and with adjacent layout optimizations. Batch or streaming writes. Auto Compaction and/or Fast Optimize. With the new incremental strategy, Liquid Clustering is highly compatible and highly efficient, and should now take its rightful place as the de facto new data layout strategy.

Read on to learn more about what Liquid Clustering is, how it works, and why you should think about it if you’re using Fabric Spark Runtime 2.0.

Leave a Comment

The Power of PBIP

Published 2026-07-28 by Kevin Feasel

Ruben Van de Voorde lays out an argument:

The PBIP format is an alternative to PBIX for storing the metadata that defines your semantic model and report. It allows more robust workflows and is far more agent-friendly, while the PBIX binary format packs everything into one file.

In this article, we argue that you should use PBIP format by default, deferring to the PBIX format only when PBIP isn’t an option.

Like when you’re using Power BI Report Server because then you’re not allowed to have nice things.

Leave a Comment

Partition Switching and Archive Tables

Published 2026-07-28 by Kevin Feasel

Aaron Bertrand shares some thoughts:

Something I’ve been thinking about lately is how much work I force onto SQL Server simply because data gets old. An order processing system may need to keep data for three years for warranty reasons or seven years for compliance reasons, but that doesn’t mean that data has to live all those years in your hottest, most active table and database. It also doesn’t mean you should spend a lot of effort moving it, row by row, to an archive table, never mind a completely different database. Only to delete it later. If you’re constantly fighting archiving processes that interfere with everyday workload, this conversation might be for you.

Click through for the article. One additional thing I could recommend looking into is using something like PolyBase for that archival. I have a demonstration of the process and it essentially uses the same technique SQL Server professionals used prior to partitioning in 2005. That particular scenario would work best when the archival data is stuff you very rarely query, but can allow you to move data off of relatively expensive fast storage onto relatively less expensive and slower storage.

Leave a Comment

Retrieving Plans from DMVs and Query Store

Published 2026-07-28 by Kevin Feasel

Deborah Melkin concludes a video series on how to get execution plans in SQL Server:

I really wanted to put this together because I feel like understanding these differences is important to understanding how we can troubleshoot performance problems and where we need to be looking for these pieces of information to get that full picture of where to spend our time. It makes us better performance tuners. I hope you found this helpful and you leave with a better appreciation for these nuances.

Click through for the video, as well as a transcription on the blog post.

Leave a Comment