Production AI and data engineering
LLM evaluation, observability, platform migrations, performance tuning, and architecture choices without vendor theater.
DWHPro Letters
The field guide for senior data engineers moving into the AI era.
Practical essays for people who still have to keep enterprise data systems correct, fast, explainable, and billable while Teradata, Snowflake, Databricks, and AI agents reshape the work.
Free forever
All articles are free. Some require a free subscription.
DWHPro Letters has no paid tier. A free subscription unlocks subscriber-only articles and lets me send the next newsletter by email, so you can stay on track instead of losing the site after one useful read.
I spend a lot of time running this site. Subscribing is the clearest signal that the work is useful and that there is real interest in future essays.
Written by Roland Wenzlofsky
Founder of DWHPro
Author of Teradata Query Performance Tuning
Market survival for senior engineers
Job-market signals, rate pressure, title inflation, and the skills that still matter when management discovers automation.
No paid version
Subscribe for free to read subscriber-only articles and receive an email when the next newsletter is available.
Latest field notes
Enterprise data engineering, AI systems, platform migration, and the labor market around senior technical work.
Featured
Latest
Optimizing Teradata Statements Containing Multiple JOINS
1. Outline
This showcase demonstrates optimizing statements with multiple JOINs using Teradata Optimizer's tuning approach. The approach efficiently determines the best JOIN strategy and implements data redistribution instead of duplication when necessary.
Identify and break down underperforming segments to optimize complex logic with multiple joins. Employ an execution
Optimizing Queries with LIKE Operator in Teradata
Learn how to optimize LIKE operator queries in Teradata to avoid full table scans and improve performance. Read on for expert tips and tricks.
Teradata Columnar Compression Methods: Run-Length, Dictionary, and Delta Compression
Learn about the benefits of compression in Teradata Columnar, including a reduction in permanent space and disk IOs. Different compression methods are used, including run-length, dictionary, and delta compression. This article explains how each method works and the advantages of using them.
Designing Small Reference Tables for Teradata: Storing All Rows on One AMP for More Efficient Queries
When designing tables for Teradata, it is important to distribute the rows across all AMPs in the system evenly. For instance, on a 100-AMP system with 100,000 rows, the objective would be to allocate roughly 1,000 rows per AMP.
I agree with the design guideline for many tables
Optimizing Teradata Stage Table Design for Faster Load Times: The Importance of Data Types and Sizes
Learn how to optimize load times in Teradata by choosing the correct data types and sizes for stage tables. Read on for a real-life example.
Teradata Statistics Design for High Performance: An Introduction
Learn how Teradata 14.10 simplifies statistics collection for better performance. Discover new features that identify unused and missing statistics.
Teradata Statistics Collection for Improved Query Performance
Learn how to optimize your Teradata queries with statistics collection. Discover what to collect and what to omit for accurate query estimation.