Press "Enter" to skip to content

Day: September 16, 2026

Tuning DiskANN

Published 2026-09-16 by Kevin Feasel

Mala Mahadevan wraps up a three-parter on how DiskANN works:

In Part 1, we saw how ‘Vamana’ represents vectors as nodes, connects them with edges, and uses greedy graph traversal to avoid comparing a query against every vector in the dataset. In Part 2, we investigated how PQ gave us compact representations for inexpensive distance calculations, caching kept useful graph nodes close, how SSDs provided capacity, and beam search allowed multiple promising nodes to be expanded together.

In this last part, we will be exploring how we use the parameters to tune DiskANN.

Click through to learn more about these parameters and what they do.

Leave a Comment

Implementing Zero-Trust for a Three-Tier Postgres App

Published 2026-09-16 by Kevin Feasel

Wellingtone Luvonga puts together some charts:

Implementing a zero-trust network model in Kubernetes requires shifting from the default-allow behavior to explicit, label-driven microsegmentation. This hands-on lab walks through securing a standard three-tier architecture (Frontend ⭢ Backend ⭢ Database) using Kubernetes NetworkPolicies, validating both ingress and egress restrictions.

Click through for the mappings.

Leave a Comment

Source Control for Power BI Semantic Models

Published 2026-09-16 by Kevin Feasel

Ruben Van de Voorde provides an overview of how source control works with Power BI reports:

Some years ago, our founder Daniel wrote a five-part blog series titled “You’re deploying it wrong!”, which was essentially him shouting that he stole fire from the software gods and handing out torches. He was talking about how software engineers had this wonderful thing called CI/CD, and how it was finally within reach of data people working with Analysis Services. Jokes aside, the core ideas have not changed, but the landscape and toolchain have, so the series deserves an update, and this is it.

Power BI has come a long way in this regard. Unless you’re using Power BI Report Server, in which case, the joke’s on you.

I’m going to return to my Power BI Report Server corner now and not have any of the nice things.

Leave a Comment

Wide Tables and Adding Columns

Published 2026-09-16 by Kevin Feasel

Richard O’Riordan runs into an oddity:

We have a wide table in MS SQL Server (2019), say 150 columns. We want to add a new bit column at the end of this table with a default value of 0, at first look at this I would have thought, that’s not a big deal, this is a meta-data only change but it wasn’t, the entire table got re-written, so here are some scenarios I tested.

This is quite odd because my recollection is that, with a default constraint, SQL Server shouldn’t need to write anything to the table, and that’s how it’s looked in my testing. Remus Rusanu also had a nice post getting into the details of how this changed for SQL Server 2012. Maybe it’s an oddity from having so many columns on the table or records of a certain width or something?

Leave a Comment