PGConf.Online 2021

High-Performance PostgreSQL

PostgreSQL is one of the leading open-source databases. Out of the box, the default PostgreSQL configuration is not tuned for any particular workload. The default configuration is designed in such a way that PostgreSQL can run on any system using minimum resources. Consequently, a default installation of PostgreSQL does not give optimum performance on the high-performance machine because it is set up to use all available resources. PostgreSQL provides mechanisms that allow you to tune your database according to your workload and machine specification. Outside of PostgreSQL, though, we can tune the Linux kernel to allow the database load to work optimally. In this talk, we will learn how to tune some of the PostgreSQL’s parameters, and we will see the effect of that tuning, but we will focus on demonstrating how to tune Linux for better Postgres performance. As there are so many Linux kernel parameters that can be tuned to improve the performance of PostgreSQL, I will also share the results of benchmarks obtained when tuning some of the Linux parameters.

Postgres Pro DBMS: what’s new & what’s on the roadmap

In this talk, you'll learn about Postgres Pro DBMS from the co-founder of Postgres Professional. Ivan will explain the philosophy behind this enhanced variant of PostgreSQL, reveal the differences between Postgres and Postgres Pro and provide the roadmap for the further evolution of Postgres Pro DBMS.

PostgreSQL partitioning. Work In Progress

Native partitioning was introduced in PostgreSQL 10, and every new release comes with more features and optimizations to this area. Yet, there is still room for improvement.

This talk briefly compares the built-in PostgreSQL partitioning with third-party extensions (pg_pathman and pg_partman), to understand what we still don't have in the core. This talk also includes an overview of partitioning-related features that are currently being developed and aimed for PostgreSQL 14.

Multicorn Foreign Data Wrapper vs plpython

Multicorn technology allows you to develop FDWs in Python, which is much easier and faster than creating FDWs in C. However, there is a downside, Multicorn FDWs work well with primitive WHERE conditions, but more complex cases cause difficulties, which I will talk about. Cases will be considered on the example of my Multicorn FDW for getting OpenStreetMap data. I will also show examples of using the same code in Multicorn FDW and plpython functions, including performance comparison. In conclusion, I will share my findings on when it is better to use plpython, and when Multicorn FDW is more preferable.