This event was conducted under the community event guidelines.

Intermediate

Basic

Conference program

10:10 – 11:05

Postgres Protocol: advantages, drawbacks, and pitfalls

Andrey Borodin

PostgreSQL сontributor, head of open-source DBMS development unit, Yandex Cloud

In this paper, I'll discuss the Postgres client-server interaction structure. Even though the protocol has been in place for 20 years, it is now evident there was a better way to implement some parts of it in the past.

Advanced

11:05 – 12:00

int –> bigint migration

When a system starts to expand, it frequently occurs that previous successful features eventually stop functioning. When it comes to data type overflow, this is precisely true. If int4 was sufficient at the start of the project, it could eventually run out entirely and you'll need to move to bigint.

Alexander Nikitin

Lead DBA, PGMechanix

In this talk, I will outline the challenges that DBAs encounter, explain how to address those issues, and present a tool that will make them much easier.

12:30 – 13:30

PostgreSQL: from a replica to a high-availability cluster

Get ready for a trip through the essential techniques for building a robust and expandable PostgreSQL cluster!

Vladimir Vinogradov

DevOps Team lead, Tantor Labs

09:30 – 10:00

Guests registration at the venue

We'll start from the fundamentals, explore the synchronous and asynchronous PostgreSQL replication, speak about Patroni, a crucial tool for managing sharding and replication, and its role in automating cluster recovery and failure detection. Using a live example, we’ll see how the cluster functions and add HAProxy and PGbouncer for more efficient load balancing and high availability, and also use keepalived to ensure continuous system operation. Generally speaking, we’ll overview a combination of technologies assemblable into a highly available and scalable PostgreSQL cluster.

12:00 – 12:30

Сoffee break

13:30 – 14:30

Lunch

14:30 – 15:20

Resolving long-standing bugs, with plv8 as an example

I'll explain how we worked around a long-standing bug in plv8 that was related to the return of scalar JSON values from Javascript functions. We managed to fix the issue, addressing it at the level of the original project to ensure stable operation for all users, rather than using crutches in our code. We will discuss whether it’s worthwhile to push this further because the format for returning NULL data still remains a problem, and also we’ll share the history of this particular pull request's acceptance on GitHub and make some suggestions for those who still doubt to get involved as contributors.

Dmitriy Dorofeyev

Chief architect, Luxms

Basic

15:20 – 16:15

PostgreSQL performance for application developers — because there is no magic button

Developing an application and stress testing is great. But as the app goes into production and begins to scale, more often than not, we begin to encounter situations that we haven't thought about during the design process. This could be related to new use cases, disruptions in technology, or the sheer scale/volume of data and users. This is especially true for databases, more so because of the increasing trend of application developers and architects defining the technology stack - i.e. non-experts of databases making database decisions. Nothing wrong with it, especially as it speeds up the process of development and go-to-market.

Umair Shahid

PostgreSQL Code of Conduct Committee Member (2019–2022) and Chair (2021-2022), Founder, Stormatics

This talk focuses on such instances in order to guide developers of a few key items to be mindful of when designing and scaling applications using PostgreSQL.

Basic

16:15 – 17:05

Step-by-step debugging of PostgreSQL source code with pg_store_plans as an example

The main focus of our study will be a thorough practical examination of the debugging process, learning to compile it using debugging symbols, and applying some fascinating techniques to find and fix mistakes. We'll cover the setup of the debug compilation environment and the configuration of VS Code for efficient debugging.

Lev Nikolaev

Developer, Tantor Labs

I'll also talk about my own journey, which began as a student with no prior work experience and ended with me being a rookie developer who found a wealth of opportunities in the database world.

17:05 – 17:25

Сoffee break

17:25 – 18:15

Wait events monitoring and low-level PostgreSQL performance issues

Wait events monitoring is a versatile method for troubleshooting performance issues; it may be used to examine any sudden degradation or insufficient CPU utilization. Performance monitoring boards in commercial products, for example Performance Insights for Amazon RDS/Aurora or the basic performance pages of Oracle Enterprise Manager, are built on wait events. I am going to show how you can get rich monitoring capabilities, comparable to Performance Insights boards, using open-source components.

Maxim Milutin

openGauss and PostgreSQL contributor

We'll dig into the types of wait events, including system (low-level) ones, and talk about mitigating the impact of factors reflecting scalability issues in the PostgreSQL kernel.

18:15 – 19:00

Debugging the PostgreSQL streaming replication

PostgreSQL replication has many intricate and fascinating aspects, making it a rather complicated aspect of the database building process. Without appropriate debugging, it is nearly hard to design this portion appropriately. To help with replication development, I'll walk you through the selection of tools and test writing techniques.

Ilya Evdokimov

Senior developer, Tantor Labs

10:00 – 10:10

PG BootCamp 2024 Minsk welcome speech

Vadim Yatsenko

Michael Goldberg

CEO, Tantor Labs

Member of PostgreSQL Code of Conduct Committee (as of 2020), Chair (as of 2023), Member of PG BootCamp Initiative Group, PGDay Israel Conferences Organizer (as of 2017)

Advanced

Intermediate

Intermediate