Sign in Subscribe

Performance Tuning

A collection of 267 issues

Why Z-Ordering Fails on Skewed Data — and Liquid Clustering Does Not

Z-ordering and Liquid Clustering both aim to improve Databricks query performance through data skipping. But when your data is skewed, one of them quietly becomes useless. A visual explanation of why — and how the Hilbert curve changes everything.

You Migrated From Teradata to Spark and Threw Away the One Thing That Made It Fast

If you have spent any amount of time working with Teradata, you know that the Primary Index is one of the most important design decisions you make. It determines how data is distributed across AMPs and whether your joins are fast or slow. Choosing the wrong Primary Index is one

Teradata Join Indexes vs. Snowflake Materialized Views — A Technical and Pragmatic Comparison

Database features should be compared based on their documented behavior, their operational impact, and the architectural principles behind them. This applies especially to physical optimization structures such as Teradata Join Indexes (JIs) and Snowflake Materialized Views (MVs)—two features often mentioned together during migration planning, yet substantially different in scope

Teradata’s Semicolon Optimization vs. Snowflake’s Architecture — Two Worlds, One Goal

Understanding Skew in Teradata and Snowflake

Why the Same UPDATE Means Something Entirely Different in Teradata and Snowflake

Teradata vs. Snowflake: Why GROUP BY Performance Differs at Scale

Fast multi-file export of Teradata query results using only Teradata SQL Assistant

I was recently approached to support a case of importing the results of a Teradata query into a third-party vendor database. On the export side, Teradata happily wrote close to 60 million rows into a single, wide CSV file. On the other side, the import process to the target database

Boost Your Teradata Performance - The Critical Role of NOT NULL Declarations

Introduction to Teradata Performance and NOT NULL Welcome to our latest Teradata performance blog post, a series designed to provide valuable insights into SQL queries. This article spotlights 'NOT NULL'. To delve deeper into 'NOT IN', it is crucial to comprehend a frequently neglected SQL database

Optimizing Teradata SQL Queries by Avoiding Full Table Scans and Utilizing Secondary Indexes

Learn how to optimize your Teradata SQL performance by leveraging secondary indexes! Avoid full table scans by bypassing the COALESCE function in the WHERE clause.