Modern PostgreSQL High Availability
It’s 3am. Your phone rings. PostgreSQL is down, you need to promote a replica to master. Why the h**l isn’t this automatic?
If you thought of this before, you want automatic High Availability (HA). Don’t miss this talk! We will enter the world of Modern PostgreSQL HA.
Good news, there are several new, “modern” solutions for PostgreSQL HA. However, there are several solutions and it's not easy to pick one. Most require non-trivial setups, and there are many small caveats about HA like how to provide entry points to the application, HA correctness, HA vs. read scaling, external dependencies, interaction with cloud environments, and so forth.
Join this talk to master PostgreSQL HA and how to deploy it on current times.
Слайды
Видео
Другие доклады
-
45 мин
Olivier Courtin DataPinkAdvanced spatial analysis with PostgreSQL, PostGIS and Python
PostGIS is well known and widely used since two decades, as the best OpenSource database solution for Spatial Analysis. This talk will focus on: spatial and advanced spatial analysis with pure PostGIS (including cutting edge PostGIS functions available); how to go further throught GeoDataScience, with Python libs and framework tied with PostgreSQL/PostGIS (including Machine and DeepLearning)
-
90 мин
Egor Rogov Postgres ProfessionalTutorial: More indexes, good and various
"And telling GIN from SP-GIST was quite beyond his wit, we found", said the classic. Can you? This masterclass is about not-so-often used index types (compared to conventional B-tree) which however can do a great job for you. We will look into internal mechanics of these indexes and discuss cases where they can be successfully applied. Also we will talk about some peculiarities of PostgreSQL index access. To spend time efficiently, listeners are required to have basic knowledge of PostgreSQL and should be used to read plans of simple queries.
Materials of the master class
Backup copy of the database with demo data can be downloaded here:
- Recovery with pg_restore (338 MB)
-
45 мин
Dmitriy Sarafannikov ЯндексHow to save statistics during major update, and what can be the consequences
It's not a secret for anyone that statistics can not be transferred with a major upgrade. For small and not heavily loaded databases this is not a problem, you can quickly collect new statistics. But we have databases with a volume of about 5TB and a load of about 100k rps, for which it became a big problem: taking off without statistics, the replicas could not even replay WAL. In my report I'll tell you what tricks we went to upgrade these databases with requirements of 100% read only availability, about what mistakes were made, and about how these errors were painfully corrected. The result of these errors was the extension called "pg_dirty_hands", in which we will collect various hacks, which can be last resort to repair data corruption.
