PgConf.Russia 2020

Kubernetes crash course for Postgres DBAs

Kubernetes is the new way of deploying software, programmatically, on almost any infrastructure (be it cloud or on-prem). But is a complex beast. How to get started? How to dive deeper? What are the specific best-practices and special hints for Postgres DBAs dealing with Kubernetes? Join this half-day tutorial to learn, practically, among other topics:

• How to quickly get started with Kubernetes
• Manage storage
• Manage services, networking and ingress/egress
• How to make Postgres cloud-native in Kubernetes
• Do a show-run of existing Postgres operators, including Zalando, CrunchyData and StackGres.

This tutorial is very practical. BYOL! (Bring Your Own Laptop). With Kubernetes installed! (check microk8s, minikube or k3s if you don’t have any installed.

Non-Relational Postgres

Postgres has always had strong support for relational storage. However, there are many cases where relational storage is either inefficient or overly restrictive. This talk shows the many ways that Postgres has expanded to support non-relational storage, specifically the ability to store and index multiple values, even unrelated ones, in a single database field. Such storage allows for greater efficiency and access simplicity, and can also avoid the negatives of entity-attribute-value (eav) storage. The talk will cover many examples of multiple-value-per-field storage, including arrays, range types, geometry, full text search, xml, json, and records.

The possibilities of profiling plpgsql code - available tools

I like stored procedures - it is great technology. But like any other technologies it allows to write not well optimized code. It is not easy to write optimized code, sql statements in complex large applications. On second hand, there are some tools, that can be used very easily, that can help. Postgres has built-in tracking functions possibility. There are PLProfiler and plpgsql_check. With these tools is easy work to detect slow part of applications.With this knowledge, the fix of performance issue is less magic.

План + запрос = ?.. Когда анализ запроса в радость

Odd things in query plan analysis - wasted time and "unnecessary" buffers.
Structural hints in a plan. How to help a developer with optimization without writing a single line of code. How to match plan nodes with query text and take advantage of this information.