© Laurenz Albe 2023 Unless you use the binary collation, creating a b-tree index to support a LIKE condition in PostgreSQL is not straightforward. This keeps surprising Oracle users, who claim that a regular b-tree index will of course always support LIKE. I decided to explore the differences between Oracle and PostgreSQL when it comes […]

There are various compelling reasons to use universally unique identifiers (UUID) as primary keys. Two examples are: To be able to generate keys independently of the database To move sets of related records between different databases without having to deal with renumbering everything However, like everything good in life, UUID’s come with their own downsides. […]

Indexing is the key to good performance. However, people often ask: Is there an alternative to btree indexing? Can we make indexes in PostgreSQL smaller? Can we create indexes in PostgreSQL faster? And how can we index in a data warehouse? This blog will answer all those questions and show which options you have to […]

Spot a missing PostgreSQL index Missing indexes are a key ingredient if you are looking for a perfect recipe to ruin performance in the most efficient way possible. 🙂  However, if you want to ensure that your database performs well and if you are generally not in favor of user complaints – better watch out […]

© Laurenz Albe 2021 Most database tables have an artificial numeric primary key, and that number is usually generated automatically using a sequence. I wrote about auto-generated primary keys in some detail in a previous article. Occasionally, gaps in these primary key sequences can occur – which might come as a surprise to you. This […]

Over the years, many of our PostgreSQL clients have asked whether it makes sense to create indexes before – or after – importing data. Does it make sense to disable indexes when bulk loading data, or is it better to keep them enabled? This is an important question for people involved in data warehousing and […]

© Laurenz Albe 2021 PostgreSQL v12 brought more efficient storage for indexes, and v13 improved that even more by adding deduplication of index entries. But Peter Geoghegan wasn’t finished! PostgreSQL v14 added “bottom-up” index entry deletion, which targets reducing unnecessary page splits, index bloat and fragmentation of heavily updated indexes. Why do we get index […]

Index missing? When an index is missing, good performance won’t be kissing a PostgreSQL user looking for efficiency but instead feels like a legacy. To satisfy a DBA’s desire and thirst, let us load some data first. pgbench is the tool of the day but the next listing will explain that anyway: Loading millions of […]

© Laurenz Albe 2020 A while ago, I wrote about B-tree improvements in v12. PostgreSQL v13, which will come out later this year, will feature index entry deduplication as an even more impressive improvement. So I thought it was time for a follow-up. Deduplication for B-tree indexes If the indexed keys for different table rows […]

Image © Laurenz Albe 2019 If you thought that the B-tree index is a technology that was perfected in the 1980s, you are mostly right. But there is still room for improvement, so PostgreSQL v12 (in the tradition of v11) has added some new features in this field. Thanks, Peter Geoghegan! In this article, I […]

A “composite index”, also known as “concatenated index” or a combined index, is an index on multiple columns in a table. Many people wonder what is more beneficial: Using separate or using composite indexes? Whenever we do training, consulting or support this question is high up on the agenda and many people keep asking this […]

  Since I only recently learned about the concept of “killed index tuples”, I thought there might be some others who are not yet familiar with this interesting PostgreSQL concept. This may give you an explanation the next time you encounter wildly varying execution times for the same execution plan of the same PostgreSQL query. […]