I'm legitimately curious -- why do people want to put EVERYTHING into postgres? I don't understand this trend (vector search, full text search, workload orchestration, queues, etc.)
I've built a number of systems that run a database and a separate search index (Elasticsearch, Solr, Xapian). The hardest part by far is keeping the search index in sync with the database. I gave a talk about this a while ago: https://simonwillison.net/2017/Aug/16/denormalized-query-eng...
Using the search engine built into PostgreSQL, MySQL or SQLite makes this problem SO MUCH less difficult.
Yes. Scale matters here. If it’s just me trying to get shit done, then one system (database, monorepo, whatever) is 100% the best. I’ve done the multiple system thing at scale, and it’s fine if I have the liquidity of engineers to do it and the cost is warranted.
Bundling everything into one system will eventually fall apart. But it’s sooo good while you can do it.
And I am decades past the point where I introduce new shit just to learn it under the guise of “needing” it. Instead I’ll introduce new things I want to learn under the guise of new things I want to learn and it will find the appropriate place (often nowhere, but you never know).
Doing everything in postgresql need not necessarily mean doing it in one database server. We could have different servers meant for search, queues, analytics (columnar) etc, all replicated (pg native replication) from a vanilla transactional postgresql server.
Now applications need only one technology, not necessarily one server.
You are right that this is the hardest, and most important, thing in search to get right. It's usually referred to as ETL. Extract, transform, load. Load: for each thing, put it somewhere for processing. Transform: for each thing process it by applying some algorithm/algorithms in one or more steps. Load: for each thing, shove it into your store. It's the transform part that is important. Extract and Load are kind of trivial to implement usually. I've seen decent implementations of only a few lines of code. Transform is application specific business logic. E and L are just simple plumbing.
What you query on is not the same as what you store in your DB. And it can be expensive to calculate and re-calculate. Especially at scale. And iterating over all your stuff can be challenging too. It requires IO, memory, CPU, etc. Your application server is the wrong place. And so is your main application database.
The challenge with search is that querying just gets a lot easier if you calculate all the expensive stuff at index time rather than at query time. Different tokenization strategies for different languages, calculating things like page rank, normalization, tokenization, semantic vectors, enriching data with other data (including denormalizing things from other data sources), etc. There are a lot of tricks you can use to make stuff easier to find.
Foregoing all of that indeed makes things simpler and faster. But your search quality will probably suffer. And if you aren't measuring that to begin with, it is probably not great. Doing all these things on write in your main database schema is going to cause other issues (slow writes, lots of schema migrations, complicated logic around CRUD, etc.). The rookie mistake with ETL is just joining the three steps into one thing that then becomes hard to run, evolve, and scale. I see that with a lot of my clients. This is textbook "doing it wrong". It's usually neither fast nor very good at search.
Even if you are going to use postgresql as your main search index, you are probably doing it wrong if your search table/schema isn't decoupled from your main application database via some ETL pipeline. That logic has to live somewhere. Even if it is a bit of a simplistic/limited "do everything on INSERT" kind of thing. That's going to hold back your search quality until you address it. There is no magic feature in postgresql that can address that. Nor in Elasticsearch (though it comes with way more features for this).
I've worked with postgresql's FTS a few times. It's pretty limited compared to Elasticsearch. Anybody running performance benchmarks should be running quality benchmarks instead. Being fast is easy if you skip all the difficult stuff. Being high quality and fast is a big challenge. And it's a lot easier with proper tools and a proper ETL pipeline.
And indeed engineering that such that the two stay in sync requires knowing how to engineer that properly. I usually start with that when I consult clients looking to level up their home grown search solutions to something a bit better.
Of course if you do ETL properly, having your DB and search index in the same place stops making sense. And if you are going to separate them, you might as well pick something more optimal for the job. There are a lot of decent solutions out there for this.
If you can avoid adding an extra service without paying too much penalty, it means not having to acquire an extra skill or hire another devops person or keep yet another service in sync / maintained / etc.
The cost of adding services to an app is so much higher than people give it credit for at organizations of every size, it's shocking to me that more care isn't done to avoid it. I certainly understand at the enterprise level that the value add of a comprehensive system is worth the cost of a few extra employees or vendors, but if you could flatten all the weird services required by all the weird systems that use them in 30,000+ employee enterprises and replace them with one database and one web/application server, you'd probably save enough money to justify having done it.
Where I work did an inventory a few years back of their systems and found that we had about the same number of databases (not tables!) as employed engineers, counting all deployed (QA and prod) instances.
The team on that inventory project obviously created a new database to put their data in, plus QA and test replicas. They (probably) have since moved to another DB system but left the old ones running for legacy applications!
We've been using Elasticsearch + PG and it's pretty nice and fast, but it adds a whole layer of extra stuff to deal with when your data is in PG but then also needs to be indexed into an external server outside of those PG transactions. In our case I'm pretty convinced it hasn't been worth the effort. I think we could've optimized PG to be as fast as we needed with a lot less overhead than dealing with an external search index.
We moved our queues to PG and it cuts out the same kind of overhead to be able to wrap an update and start a job in a transaction. PG has been plenty fast to keep up with our queue demand.
Ultimately I think being able to do things transactionally just avoids a whole class of syncing issues, which are basically caching issues, and cache invalidation is one of the 2 hard things.
You can abstract this to any RDBMS, and the justification is that it makes everything a lot faster & easier.
I just got off a call with a client where their developers were using ORM-style abstractions to manipulate data for downstream processing in code, turning what should have been a few seconds of one custom SQL command into several hours of passing objects around multiple computer systems.
If we can put the FTS engine inside the SQL engine, we can avoid the entire space of APIs, frameworks, 3rd parties, latency, etc that goes along with otherwise integrating this feature.
Modern SQL dialects represent universal computation over arguably the best way we know how to structure complex data. Adding custom functions & modules into the mix is mostly just syntactic sugar over what is already available.
There is ZERO honor in offloading what could have been a SQL view into a shitty pile of nested iterators somewhere. I don't understand where any of this energy comes from. The less code the better. It's pure downside.
> There is ZERO honor in offloading what could have been a SQL view into a shitty pile of nested iterators somewhere. I don't understand where any of this energy comes from. The less code the better. It's pure downside.
I wholeheartedly agree with you. As to why we use ORMs, the impression I get from the engineers I work with is that many of them a) don’t know SQL and b) feel like it’s “data analyst” stuff and so beneath them to learn it. Real engineering requires objects and inheritance or structs, pointers and arrays (depending on the engineer).
I think it’s the declarative nature of SQL that turns them off.
I've had to experience it firsthand again a while ago but yeah.
I was replacing an application management interface of sorts, large ish sets of configuration parameters, ideal for a relational database. But I wanted to treat the combined configuration as a document, since that's what the front-end would send over. Ended up using GORM, which was fine for a little while... but quickly falls apart, especially when your data model is nested more than one level deep. And then you end up having to figure out "how do I solve X in GORM" and find yourself with limited documentation and a relatively small community whose members quickly burn out of trying to help people.
Avoiding distributed systems problems. Distributed systems are so incredibly hard to get right that I will vertically scale postgres until I hit an insurmountable wall before giving in.
You can also build distributed DBs with PG. For example for a DB with multiple write nodes all you need to do is implement an event shipping model with logical replication where your servers publish their events and subscribe to others, and you need to implement conflict resolution rules, naturally. I think PG's type system can probably be leveraged to make a CRDT system on top (and I bet someone's already done it).
IBM mainframes were created for you. Imagine you had a single computer that had multiple nines reliability. Hot swappable disk, RAM, CPU. Redundant power supply. Redundant network stack. OS designed to never need restarting. That's basically what a mainframe is, and IBM sells billions of dollars worth of them to this day.
It's because postgres is in fact good at a lot of vaguely database-looking things. Even if it weren't the best for anything, if it does 80% of things at 80% best possible — it is reasonable to have postgres as "first thing to reach for" by default.
That said, it's easy to forget to check if you're in either of those 20% (or both.) There's probably a whole bunch of postgres usage where really something else should be used, and people just never checked.
Because PG is a fantastic platform for doing everything you need in SQL and it performs real well. Add PostgREST and you've got a REST API almost for free (you have to design a schema of what to expose, so not entirely free, but still, pretty close). Also, PG moves _fast_ and has a very active developer and user community, which means you'll be getting more awesome functionality in the coming future. (E.g., I've been following the AIO thread and the performance improvements coming from that patch set will be hefty.)
When you're starting something new and the amounts of data is still small, it's often better early on to focus on the product than optimizing for theoretical performance optimization that may never pan out (either because the project will fail or that the bottlenecks may ultimately not be what you thought or expected).
At my current gig, we used to shove everything (including binary data) into postgres because it was easy and all our code plugged into it anyways. When it started to become uneconomical (mostly due to RDS storage costs), we then started shunting data to S3, DynamoDB, etc.
Also, not everybody can be on a cloud with easy access to all the fancy products for queuing, caching, etc. Sometimes it's better overall to have to deal with one complex beast (that you'd have to maintain anyways) than spending time deploying Kafka, MongoDB, etc (even though it can sometimes be easier than ever with pre-built manifests for K8s) as well as securing and keeping them all up to date.
I do strongly encourage people to treat code that deals with these things with as much abstraction as possible to make migrations easier later on, though.
Tbf everything you mentioned, MySQL also has (in the latest version, for vectors).
But either way, the answer is simplicity and cost. I assume you’ve heard of Choose Boring Technology [0]? Postgres is boring. It’s immensely complex when you dive into it, but in return, you get immense performance, reliability, and flexibility. You only have to read one manual. You only have to know one language (beyond your app, anyway – though if you do write an app in pure SQL, my hat is off to you). You only have to learn the ins and outs of one thing (again, other than your app). ES is hard to administer; I’ve done it. Postgres is also hard to administer, but if I have to pick between mastering one hard thing and being paged for it, or two hard things and getting paged for them, I’ll take one every day.
It is very good at doing the job that people over-eagerly offload to specialized services. Queues, notifications, scheduled jobs. And can be specialized with extensions.
I am going to hazard a guess that it's because the closer the services are to your data, the easier they are to implement, and you often get great speed too. FTS in Postgres has been fantastic for me, and combining it with vector search and RAG gives you a pretty sweet deal for low effort.
Disclaimer, I have no experience with this kind of thing. However, theoretically, less tools is better for an organization - see [0] - and if your job adverts say just "postgres" instead of "postgres, elasticsearch, tool x, tool y, tool z" etc, you don't need to find (or train) a unicorn that is up to speed on all of them.
That said, "postgres" is a very broad subject if you take all of those into consideration, if you need to specialize your search for someone who knows how to do X in PG specifically you're almost back at the same spot. (I say almost because I'm sure it's easier to learn a specialization in Postgres if you're already familiar with Postgres than it is to learn a completely new tool)
And caveat, there's a high golden hammer risk there. I'd start questioning things when needing to query JSON blobs inside a database.
I agree with one of the points from The Art of PostgreSQL by Dimitri Fontaine, which is appropriate for answering this question:
"So when designing your software architecture, think about PostgreSQL NOT as storage layer, but rather as a concurrent data access service. This service is capable of handling data processing."
All these replies have me so confused, the reason to shove everything into your database when you can is because you can transact across them. That's thing you can't get once you have a second system.
It’s free, it’s stable, it’s excellent, and it’s everywhere. The question should be “why don’t people put everything in Pg?”
The only things you shouldn’t put in Pg are things where there’s an obviously better alternative that’s so much better as to make running an extra service worth it. There definitely are cases where that’s true (Redis when you need exceptionally quick responses, for example) but it’s a high bar to clear.
Because the vast majority of people on HN or in the real world don’t need to scale beyond 10 concurrent users. It’s insane how much infrastructure script kiddies add to their projects when it could be done in the database and scale well.
Sorry you’re being downvoted; you are correct. I love Postgres, but devs absolutely flock to it because influencers said to. At a job a while ago, my team put out a poll asking for devs opinions and reasons for their preferred RDBMS. Every single one said Postgres, but no one could elaborate as to why. One said “it’s more flexible,” which is true, but no one there was using ANY of its flexibility.
That’s the part that baffles me. You’ve selected a DB with native support for esoteric but useful data types like INET (stop storing IP addresses as strings in dotted quad!), and a whole host of index types beyond B+tree, but they’re never using them.
Read your RDBMS docs, people. They’re full of interesting tidbits.
I think you're looking at it from a weird angle...
"Every single one" in your team agreeing on a single specific technological choice is one of the rarest things I can image! Developers argue about libraries, frameworks, programming languages, services, etc., and I think it speaks for itself if Postgres is the thing that comes closest in bridging the gap at least on one layer in the tech stack. Postgres is a "conservative" choice with a very active community and extensible ecosystem.
Also, nobody is ever making use of their technological choice to its full extent, you'd rarely know what you'll need beforehand, and it's just nice not having to add other storage engines when that one feature request steps into your life.
