Press "Enter" to skip to content

Day: June 8, 2026

A Primer on Apache Iceberg

Published 2026-06-08 by Kevin Feasel

Walt Riberio lays out an explanation:

Data lakes had a reputation problem. The promise was compelling: dump all your data into cheap object storage—S3, GCS, Azure Blob—and query it whenever you need. The reality was a mess of stale partitions, schema drift, and silent data corruption caused by unsafe concurrent writes. Engineers knew the risks and worked around them rather than fixing them.

Apache Iceberg was built to fix that. And it’s catching on fast—even in the Kafka® world Iceberg is a hot topic. If you’re running ClickHouse® or building a pipeline that feeds into it, Iceberg is quickly becoming hard to ignore if you care about accuracy at scale. It’s becoming the connective tissue of the modern data stack.

Click through to learn more about the product and where it potentially fits in an organization.

Leave a Comment

The Ulam Prime Spiral

Published 2026-06-08 by Kevin Feasel

Tomaz Kastrun re-creates a classic:

Stanislaw Ulam, Los Alamos, 1963 was bored in a meeting and he started dooddling integers in a spiral and circled the primes. Diagonal lines appeared. He later showed it to Martin Gardner, to Ulam surprise, Gardner published his findings in Scientific American. We are still confused to this day.

Click through for a demonstration of this in action.

Leave a Comment

What pg_stat_statements Cannot Do

Published 2026-06-08 by Kevin Feasel

Radim Marek enumerates a list:

Part one made the core case: pg_stat_statements counts, it doesn’t record. It walked through how the queryid jumble fragments one logical query into many rows, how the first-seen text freezes your per-request tags, and how the averages bury the p99 that actually pages you. All of that was about data the extension has and distorts.

This part is about the rest: the entries it silently throws away, the query text that can vanish all at once, the plans and replicas it never records, and the knobs that bite. It ends where part one started, with the question the whole investigation was really about: is this the query store Postgres is missing, or just the floor you’d build one on?

Click through for those limitations.

Leave a Comment

The Limits of Parallelism with Fabric Dataflows Gen2

Published 2026-06-08 by Kevin Feasel

Chris Webb notes that parallelism does not mean “free performance improvements”:

To finish off my series of posts on concurrent evaluation in Fabric Dataflows Gen2 (see part 1 and part 2) I decided to do some more realistic tests to see how much parallelism I could get. To do this I uploaded 244 identical Excel files containing almost 542000 rows of data each to a SharePoint document library. Excel files are probably the worst-performing file format for dataflows (see here for some tests that show this), while SharePoint is probably the worst-performing place to store data for a dataflow and also has a reputation for throttling applications that make too many requests.

Click through for Chris’s test results.

Leave a Comment

Optional Parameter Plan Optimization

Published 2026-06-08 by Kevin Feasel

Hugo Kornelis concludes a mini-series:

It’s time to finish the triptych on bad parameter sniffing, and how Microsoft tries (and fails) to fix this for you.

After first talking about bad parameter sniffing in general, I used my last video to explain Parameter Sensitive Plan Optimization, the feature that Microsoft released in SQL Server 2022 as an attempt to fix one of the three root causes for bad parameter sniffing. If you want to see why I consider that a failed attempt, go watch that video.

In SQL Server 2025, Microsoft then added Optional Parameter Plan Optimization (OPPO). Their attempt to fix the second of those three root causes. Put where PSPO in most cases simply doesn’t do anything, or does do something, but not enough, OPPO does too much. It fixes bad parameter sniffing due to optional parameters in demos that are carefully curated to showcase the feature, but it does more bad than good in more realistic scenarios.

Read on for Hugo’s overview, as well as a video.

Leave a Comment