Have you ever tried to delete a few million rows from a table with several hundred milllions of rows with a DELETE statement? If yes, then you know that this is a very bad idea. With Oracle Partitioning you can do this more elegant and much faster.
PL/SQL functions in SQL statements may cause major performance problems. If it is not possible to avoid these calls, there are several solutions to improve the performance. But not for all situations.
A typical reason of many performance issues with SQL are function calls in the WHERE condition. They make it difficult for the query optimizer to determine the cardinality. Fortunately, there are several ways to solve this problem.
One of the extensions in Oracle 20c is the possiblity to use the In-Memory Database option for Partitioned External Tables and Hybrid Partitioned Tables. In my opinion, this opens up many possibilities to perform efficient ad-hoc queries on Data Lakes. That’s why I prepared a demo script for my DOAG presentation about SQL features in Oracle 20c. Unfortunately, it turned out differently than planned. A drama in four acts.
I work with SQL since decades, and I think to have some practical experience with writing SQL queries on Oracle databases. So why should I buy a book about practical Oracle SQL? If you are in a similar situation like me, read this book review.
Although Query Rewrite on Materialized Views is a powerful feature which is available in Oracle databases for a long time, it is rarely used. Many developers and database administrators don’t know about it or do not use it because they think it is too complex. In this blog post, I try to reduce the concerns about complexity with some general design tips for Query Rewrite.
If you work with Data Vault for a data warehouse running in an Oracle database, I strongly recommend to use Oracle 12.2 or higher. Why that? Since Oracle 12c Release 2, join elimination works for more than one join column. This is essential for queries on a Data Vault schema.
NLS (National Language Support) parameters are very useful for multilingual applications in an Oracle database. But it is important to understand their impact on query performance. I was aware that NLS is important for indexes, but never thought about its impact on partitioning.
Oracle 19c provides a set of new SQL functions for bitmap operations. They allow fast computation of COUNT(DISTINCT) expressions. What is the purpose of these functions? The Oracle documentation is very sparse, but at least it tells us that the main focus are Materialized Views on aggregated data. Let’s have a detailed look at this new feature.
Materialized Views are often used in data warehouses to improve query performance on aggregated data. But what if it takes too long to refresh the materialized views? Here are some basic rules to improve refresh performance.
Data Warehousing with Oracle 