One way the Postgres project is subtly misleading, is that it becomes easy to forget that not all other projects are nearly as well managed. This becomes more relevant when delving into niches that lack sufficient visibility to expose the more obvious deficiencies. As much as we like Postgres, it’s not quite as popular as it could be. This makes some of the side projects infrequently used, and as a direct consequence, they can often resemble jerky automatons cobbled together out of spit and bailing wire.
Items in Category: Tech Talk
There comes a day in every young database’s life that it’s time to move on. I’m sorry 9.4, but the day has come that we must say goodbye. It’s not like we haven’t had our [intlink id='pg-phriday-high-availability-through-delayed-replication']good times[/intlink]. While I truly appreciate everything you’ve [intlink id='pg-phriday-materialized-views-revisited']done for me[/intlink], we must part ways. I’m far too needy, and I can’t demand so much of you in good conscience. May your future patches make you and your other suitors happy!
Ah, source control. From Subversion to git and everything in between, we all love to manage our code. The ability to quickly branch from an existing base is incredibly important to exploring and potentially abandoning divergent code paths. One often overlooked Postgres feature is the template database. At first glance, it’s just a way to ensure newly created databases contain some base functionality without having to bootstrap every time, but it’s so much more than that.
Now that we’ve decided to really start embracing horizontal scaling builds, there is a critically important engine-agnostic element we need to examine. Given an existing table, how exactly should we split up the contents across our various nodes during the conversion process? Generally this is done by selecting a specific column and applying some kind […]
I always advocate breaking up large Postgres tables for a few reasons. Beyond query performance concerns, maintaining one monolithic structure is always more time consuming and consequentially more dangerous. The time required to create a dozen small indexes may be slightly longer than a single larger one, but we can treat the smaller indexes as incremental. If we want to rebuild, add more indexes, or fix any corruption, why advocate an all-or-nothing proposition? Deleting from one large table will be positively glacial compared to simply dropping an entire expired partition. The list just goes on and on.