PGConf.Russia 2023

The following topics will be covered as part of the presentation:

• Architecture of logical replication
• Publisher introduction
• Subscriber introduction
• Data syncronization introduction
• Logical decoding
• Replication slot
• output plugin

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.

I'll show in PostGIS how everyone can analyze the geodata of the entire Earth and get answers to their global questions in minutes and seconds.

When you use a taxi in small towns, calling a car by phone, then with a high probability your trip is charged by the program based on OSM data. For billing, one of the routing packages is used. Through this use case, the taxi company employees put the house number and street on the buildings and contribute not only to their business, but also to OpenStreetMap.

The data analytics scenario also includes the tasks of where it is better to place an outlet so that buyers come to it. Again, data on walking distance and population of the surroundings can be extracted from geodata. You can calculate the value of real estate based on many factors related to the location of the object and its surroundings.

Scientists can build predictive models to predict epidemics, urban evolution, plan recreational areas and development of existing territories based on open geodata.

Well, you can answer any geography question that comes to your mind: calculate the area of cities and buildings, the length of roads and extract the names of cities, regions and islands. You can, for example, become a champion in the game of "Cities" or establish a new service for renting electric scooters. Everything is limited only by your imagination.

I published https://github.com/igor-suhorukov/openstreetmap_h3 - my project of high performance data loader, which allows you to perform geoanalytics on data from OpenStreetMap in PostGIS. It transform OpenStreetMap World/Region PBF dump into partitioned by H3 regions schema. Columnar storage option activate CitusDB extension in PostgreSQL to speedup aggregation queries.

With the rise of NoSQL databases, a number of falsehoods have flourished regarding how to choose a database engine. This talk focuses specifically on Redis and PostgreSQL, and why one might choose one or the other.

At small scales, we can often get by thinking of database servers as black boxes, but as we scale, the internals and architecture become more and more important. This talk focuses on behavior of these systems at scale and under load.

In this presentation you will learn:

• How Redis and PostgreSQL differ architecturally
• How differences in architecture affect scalability and performance
• Cases here Redis is the clear winner
• Cases where PostgreSQL is the clear winner

Additionally, some notes will be offered in terms of where PostgreSQL can improve in to compete with the sorts of workloads that generally favor Redis.