I’ve been reading up and here are some thoughts.



PgBouncer internals

  1. Client connects to PbBouncer.
  2. Client makes SQL request / query / transaction
  3. Gets a response.
  4. Repeat steps 2–3…



Today’s post is about real life of PG’s Write-Ahead log.

WAL. An almost short introduction




Real life with over 100 PG clusters in different regions.

Topology

  • has ~125 OLTP database hosts, 94% of it is PostgreSQL
  • Some of the old and all the new are RDS
  • One cluster (the oldest) serves ~300,000 transactions per second
  • i2.8xlarge AWS instances allow to keep write latency down
  • uses Auto Scaling Groups for read replicas


Как отодвинуть проблему недостаточного размера Posgres RDS-инстанса и сэкономить деньги.

  • Не секрет, что Postgres очень не любит большое количество одновременных соединений.
  • У нас пришла пора апгрейда инстанса в связи с нехваткой памяти из-за over 1k одновременных подключений к БД.
  • Docker образ с PgBouncer.
  • На базе него — сервис в ECS-кластере.
  • Для большей производительности мы используем "networkMode": "host". Однако, внимание —…


Why high # of active connections is not a problem, and what to do with waiting client's connections.

Monitoring PgBouncer

pavel trukhanov

Founder at okmeter.io monitoring

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store