PGConf.Russia 2023

A report on our experience of implementing PostgreSQL instead of Oralce in a highly loaded system with a microservice architecture that processes several terabytes of data. Let's talk about how we went from piloting Greenplum to moving to several interacting PosgreSQL databases with different load profiles, and about the problems we encountered in the process.

In my opinion, every DBA has to deal with "heavy" queries. The question is whether we can do something at all to speed up such query. Is it possible to optimize such queries, or maybe it doesn't make sense? For example, rewriting such query may take much longer than getting any performance gain from this rewritten query. During this talk, I'll tackle several approaches to speeding up queries, and demonstrate some practical examples of optimizations that I use at work.

How does network latency affect performance when using synchronous replication? If a primary node fails, how long will our cluster not be reachable? How much data can be lost? Will there be any anomalies when reading from standby? We are designing our network architecture, give us a minimum RTT value for cluster solution! Many DBAs have to face these issues when designing and implementing cluster solutions. In this short talk we will tell how to quickly estimate RTT, RTO, RPO values, with the help of which Open Source tools one can test cluster solution and experimentally verify the design parameters, which metrics values can be used as a guide in design.

There is set of Ansible roles for any Postgres configuration: a standalone instance; a streaming replica; patroni+etcd; stolon+etcd; patroni+consul; stolon+consul. In this talk, we will discuss why using this set of Ansible roles is better than other solutions.