Make Your PostgreSQL 10x Faster on Cloud in Minutes

Cloud storage has some unique characteristics compared to traditional storage mainly because it is virtualized and controlled by software. One example is that AWS EBS shows higher throughput with larger I/O size up to 256 KiB without hurting latency. Hence, a user can get only about 4 MiB/sec with 1,000 IOPS EBS volume if the I/O request size is 4 KiB, whereas a user can get about 250 MiB/sec if the I/O request size is 256 KiB. This is because EBS consumes one I/O in a given IOPS budget for every I/O request regardless of the I/O size (up to 256 KiB). Unfortunately, PostgreSQL cannot exploit the full potential of cloud storage because PostgreSQL has designed without considering the unique characteristics of cloud storage.

In this talk, I will introduce the AppOS extension that improves the throughput of a write-intensive workload by 10x by transparently making PostgreSQL cloud storage-native. AppOS works like a storage driver that efficiently exploits the characteristics of cloud storage, such as I/O size dependency to storage throughput and latency, atomic write support in cloud block storage, and fast, but non-durable local SSDs. To do this, AppOS comprises a Linux-compatible file I/O stack including virtual file system, page cache, block I/O layer, cloud storage driver. On top of the file I/O stack, syscall module supports registering pre- and post-handler for file I/O-related system calls in order to transparently work without modifying PostgreSQL codes.

I will focus on presenting key use cases and performance results of the AppOS extension after explaining the internals. Specifically, I will show the performance results of OLTP and some batch workloads using standard benchmarking tools like pgbench and sysbench. I will also present performance results and implications on multiple clouds including AWS, GCP, and Azure.

StackGres: Cloud-Native PostgreSQL on Kubernetes

An enterprise-grade PostgreSQL requires many complementary technologies to the database core: high availability and automated failover, monitoring and alerting, centralized logging, connection pooling, etc. That is, a stack of components around PostgreSQL. Kubernetes has enabled a new model to deploy software abstracting away the infrastructure. However, containers are not lightweight VMs, and the packing of software paradigms that work on VMs are not valid on containers/Kubernetes. How should be PostgreSQL and its stack be deployed on Kubernetes? Enter StackGres. An open source software that is the result of re-engineering PostgreSQL to become cloud native. Join this talk to learn and see demos of how to generate PostgreSQL minimal containers; how the sidecar pattern is used (abused) to integrate PostgreSQL’s stack components, and how the networking and storage are handled. More info: stackgres.io.

Managing PostgreSQL at Gitlab.com

I would like to present the main projects for the evolution of our database, how we execute the administration, the problems and pitfalls we found, and how we solve them,the number and how are  the database clusters from Gitlab.com , and what is our planning for the future, sharding, kubernetes... Our environment is in an exponential growth, with millions of users and thousands of requests per second, and we keep our platform stable and scaling. Join our session and discover our how we are doing it!

Инструмент как код: тестируем Postgres

На мастер-классе научимся проводить быстрые нагрузочные тесты баз данных Postgres: оптимизаций конфигов базы, структуры данных, индексов, настроек ОС и т. д. Для этого создадим код, из него поднимем инфраструктуру для теста и проведём сам тест. В результате получим гибкий инструмент в коде, к которому можно прикрутить любой мониторинг и за который не надо платить большие деньги, т.к стенд создаётся за 7 минут в пустой AWS учётке и убивается после проведения тестов. Для этого мастер-класса есть важная подготовка, которую нужно сделать заранее, чтобы в полной мере попробовать все, что спикер хочет предложить. Один шаг не быстрый - нужно сделать триальную учетную запись в AWS. Для этого требуется подтверждение регистрации от Amazon, которое они делают в течении 24 часов (если вы ранее работали в AWS, и у вас есть учетка - это хорошо, если нет - нужно пройти этот путь) Также, лучше заранее поставить последние версии ansible и terraform.