In this presentation I will try to analyze the experience of using pgbouncer and other database proxies at prom.ua. What profit does it bring in organizational sense, what new flexibility it gives to the infrastructure and everyday maintenance and scaling, which additional enchancements we managed to implement and test, what worked and what did not.

In this presentation I will try to analyze the experience of using pgbouncer and other database proxies at prom.ua. What profit does it bring in organizational sense, what new flexibility it gives to the infrastructure and everyday maintenance and scaling, which additional enchancements we managed to implement and test, what worked and what did not.

During the lecture, Mykyta will share the story of Solidgate's journey in building a high-performing and reliable fintech company, striving for 99.999% uptime on AWS’s SaaS platform. He'll uncover numerous caveats in doing things right without full system access, addressing product requirements, and staying up to date.

Imagine you are designing a B2B service that will serve millions of businesses. This service will have dozens of different microservices with their own data, which can contain millions of records. How do you design such a database? Why is sharding not always the answer? What other options are there for such an architectural solution? I'll tell you how we at Uspacy came to serve thousands of small databases instead of a few large ones, what we've encountered and what we plan to face)