In Praise of PostgreSQL
drewdevault.com
drewdevault.com
I realize that this isn't the main point, but that aside strikes me as high praise for SQLite — which has also always struck me as exceptionally well built software.
SQLite occupies the domain of embedded (possibly in-memory and on low-resource systems) databases in which the subset of SQL unrelated to access control is all that's needed. Think of databases that take the place of application configuration, your browser bookmarks and cookies, a music library database.
They are both excellent pieces of software, yes, but they don't even live in the same problem space despite both having "SQL" in their names.
(I highly recommend to watch it, if only to get a sense of the smart and funny energy that Dr Richard Hipp puts off!)
That's true in theory, but I do find that they end up competing in practice when you're looking for something in between the two extremes.
I often seem to find myself in a position where I need a database, but it's only for one day's data, which is not huge but not tiny (say 100's of MB). It may be helpful to have a couple of processes accessing it, but probably only one will be a writer, and it's very likely (but not certain) that those processes will be on the same computer. It useful to have that data in a single file for archiving. Those are situations where you currently face a geniune choice between PostgreSQL and SQLite - more because both of them are a bit wrong, rather than because they're both perfect for it.
I wish there was something in the middle: a separate program you could spin up, so it's a client-server database like PostgreSQL, but it's just a single binary that doesn't need installing. And then it operates on a single file that it creates on the fly (or can open an existing one of course), a lot like SQLite. Hmm, as I think about it, it actually wouldn't be hard to make program that does that by making a thin wrapper around SQLite with e.g. gRPC.
I would definitely not try using this in a write heavy workload but could be interesting to consider in a write occasionally scenario.
At the time I was able to push it into something like 10 simultaneous write accesses (transparently serialized by SQLite). It was an adaptation of a MS Access file that would fail at approximately 5 people opening it at the same time (independently of their actual usage). Luckily I got access to a Postgres installation shortly afterwards so I didn't have to push SQLite any further.
I think it's a pretty different use case to what the parent wanted
Exact same story for its HyperSQL and Derby friends: http://hsqldb.org/doc/2.0/guide/running-chapt.html#rgc_hsqld... https://db.apache.org/derby/#What+is+Apache+Derby%3F
Redis is considered an in memory store as well, even though it can persist it's data on disk too. And the capability isn't what makes it in-memory either, as you can store a SQLite db only in-memory too, and nobody would consider it to be an in-memory database. You could technically do the same for PostgreSQL, though that's not officially supported and needs a little effort on the users part.
I guess my phrasing was poor if you misunderstood me there, but I do stand by what I said: while H2 is a great project in Java-Land, it's very different to what I understood the parents goal to be.
The main consideration is typically about how the database is used. "Probably only one will be a writer" is a strong argument in favor of SQLite, which doesn't really have much of a concurrency story for writes (multiple processes can open the same SQLite database for writing, but they implement a block so only one at a time can actually write).
The file locks can be tricky to get working over NFS/CIFS with multiple computers, but it's still possible.
It's intended to be and is deserved.
It’s lightweight, reliable, knows what it’s tasked for, does exactly wat it does, and easy to extend. I personally reach SQLite first in so many cases, including web backends. That the DB is a file that’s literally a cp away to backup is super-useful in development and debugging.
Really the only complaint is that type enforcement is lacking… which I wish I could have said not a big deal, but it turns out it is. I wish I had a embedded DB that’s as reliable and trustful, lightful as SQLite, but with types enforced. :-(
But this is worth considering as an option.
Firebird can be used as an embedded DB or an client/server DB. It supports a wide variety of data types, is open source and cross-platform too.
It's a database that's rarely discussed here (a mystery why not). It certainly deserves wider attention.
Firebird features: https://firebirdsql.org/en/features/
I know it’s completely unreasonable to judge Firebird today for Interbase 20 years ago, but admit that’s still the first thing that pops to my mind.
Do you feel Firebird offers something SQLite doesn’t?
Firebird can be used as a proper client/server database engine. It supports static data types and stored procedures (unlike SQLite). It sits somewhere between SQLite and PostgreSQL. So it may be worth evaluating if you don't want the heft of PostgreSQL but need features missing from SQLite.
Sounds like it's not quite as embedded as SQLite? Seems like it just smushes the server into your app and spins that up from within. Also it only seems to be properly "embedded" on Windows (no macOS support and Linux is a bit weird?)
> Finally, you can't just ship libfbembed.so with your application and use it to connect to local databases. Under Linux, you always need a properly installed server, be it Classic or Super.
I agree it could be better packaged, but it's actually not impossible to ship a Postgres with your program.
What do you mean? It's not like postgres is bundled by default with most distros.
Yes, the OS doesn't impose any opinion. It's just a matter of what application developers do, and the developers of Linux applications tend to write code that doesn't require installing. As an example, there's Postgres :)
There is nothing about linux that makes postgres more or less portable than on windows or osx, and I'd be interested if you have any examples of postgres being used in a non-packaged way (outside of development tools like postgresapp.com).
Besides that, one of the strengths of most linux distros is that they provide a centralized way to install/upgrade packages like apt/yum/dnf/pacman and don't have to rely on third party tools like homebrew.
WRT "embeddable postgres":
Postgres relies on having multiple processes that collaborate on making database access fast. And it requires there to be only a single instance of postgres that can access the data. There's some inherent increase in difficulty of embedding something like that compared to something with sqlite's architecture.
Until recently PG didn't have a way to provide non-tcp access on windows, which made embedding on windows a bit more problematic. But since Win 10 unix domain sockets are available on windows, so things have gotten better.
I'd guess that after that the fact that a postgres installation consists out of many files, instead of a shared library or two, is the biggest difficulty. Those files are relocatable at least, but it's still far less convenient.
I don't know how much of a relevant factor the layout of the "data directory" is - for some database embedding scenarios it sure is convenient to only have to deal with a file or two. But moving a directory around isn't that much harder...
My guess is that somebody with interest could improve the situation measurably within a reasonable timeframe...
1. Postgres doesn't really didn't want to be statically compiled, according to [1].
2. Messing with LD_LIBRARY_PATH to point to an embedded glibc and openssl when running the heremetic postgres binaries.
3. Postgres will absolutely refuse to run the server process as root. That's sensible for a production environment but a pain in CI because it would require setting up another user on machines. I patched away the problem to allow Postgres to run as root.
On the bright side, with some minor tweaking, Postgres can start a fresh database in 300 ms which is plenty fast for tests and you avoid spinning up Docker containers. I ended up using template databases to only spin up the database once per test suite. Each tests then copies the database template for each test in the suite which reduced the setup time per test to ~20 ms.
Instead of using rules_foreign_cc to build Postgres from source with Bazel, I ended up building Postgres outside of Bazel with Docker and zipping it up by target platform (linux|darwin)_(arm64|amd64).
https://www.postgresql.org/message-id/4E0DE1B6.6070600@postn...
I'm guessing it's not open source - as I couldn't find it on your Github account or your company's account. If that's the case, would you be able to create a quick gist with a copy/paste of any of the code you can share? If you have the time I'd appreciate it!
Related links for anyone interested: - dataform uses bazel for ci tests. They builds redis from source, but run postgres as a container. See here: https://www.reddit.com/r/bazel/comments/kcmbwb/how_to_run_se...
Right. There's a lot of features that'll flat out not work if you somehow compile it statically. I don't think it's a useful thing to run tests with a version of postgres built like that. Nor does it really solve anything - postgres will still require a bunch of on-disk files.
You shouldn't need have to do so anyway - you can relocate the postgres installation itself when compiled normally. Binaries like postgres will try to find the files they need relative to their own location if not found at the builtin location.
Of course, libraries that PG binaries dynamically link to need to be somewhere in the library path. If you need to adjust that you can do it by adding rpaths to the binaries/libraries to additional locations, so you don't need to adjust LD_LIBRARY_PATH.
Oracle has a lot of advanced functionality that no other RDBMS really has, and I do hit sharp edges of PG (especially in unexpected query planning decisions) more than I do in SQL Server.
And, as much as I hate MySQL, I have a few instances that are approaching 10 years old with billions of records that have almost never needed query optimization. I just added indices in logical places and they just work somehow.
For example https://www.postgresql.org/docs/13/wal-configuration.html is done properly: it describes what that part of the program actually does, with the settings that affect it introduced at the appropriate point, separately from the reference documentation which lists the same settings in alphabetical order.
But not all aspects of Postgres get this quality of documentation. It would be good to see a similar treatment of the checks the planner or executor performs to decide whether to permit access to a particular table and column and row, rather than having to deduce it from the separate descriptions of the security features.
(For example, the basic fact that a VIEW accesses a table using the privileges of the view's owner, not those of the user executing the query, is very hard to find in the documentation.)
Interestingly enough the competition for "biggest impact on my SQL Knowledge"... Is a book about MS Access 97 which taught me enough about how SQL actually works that I could understand Postgres docs.
For most companies I've worked at I've used either SQL Server or Oracle and from what I've seen when looking at job posts recently the demand is only increasing. Although I live in Los Angeles and not Silicon Valley.
Additionally Facebook and other large tech companies still use MySQL and outside of silicon valley many large companies still run IBM databases.
I don't see the other major databases being obsolete at all.
I've used PostgresSQL before for learning (~3 years ago) but didn't really like the web based UI compared to SQL Server or Oracle query tools.
Does anyone have recommended PostgresSQL query interfaces?
SQL Server is really expensive. But, the more Microsoft stack you have in house the more discounts you get. The more Microsoft certified staff you have, the more discounts you get. This biases companies to hiring more developers with Microsoft certs, which leads to more developers getting Microsoft certs, which leads to more discounts and a feeling of investment at higher levels of the company that lead them to be unwilling to switch.
I spent 5 years at a company that was using SQL Server as the backbone of it's flagship product. They pushed it as far as they could, but ultimately the biggest issues with the product at the company revolved around either limitations of SQL Server, lack of features with SQL Server or licensing being prohibitive of different types of beneficial scaling strategies. I pushed for them to make the switch to PostgreSQL for probably my last 3 years there as part of the architecture group, including migration and training plans, etc but it was like talking to a wall. Nobody could acknowledge the SQL Server issues. Everyone who had a part in considering solutions was too invested as far as I could tell. Eventually they tried moving to Couchbase (NoSQL) before finally making the switch to PostgreSQL from what I heard from people I know who still work there.
As far as query interfaces go, you can use just about any tool you like. I prefer Navicat myself but I know a lot of people like DBeaver and DataGrip.
I've been using Navicat for SQLite and like it. If I ever get a PostgresSQL job I'll try it there since it works well for SQLite.
At my current job we are on IBM AS/400 for ERP (legacy and replacing it would be very high risk as it would change the business drastically so that won't happy any time soon).
For all other stuff we are using SQL Server and several of our key 3rd party vendors only use SQL Server. I think the fact too that some many companies run Microsoft Network and Servers that it naturally lends toward the network effect as you mention. Plus while it's expensive it's 10x cheaper than Oracle in many cases.
I don't use Web DB UIs, but I like dbeaver and pgcli a lot. You may find them useful.
I think for Enterprise SQL Server is increasing if anything (due in part because so many Companies run windows Servers internally) and developers can install it for free.
I do like the power of PL/pgSQL although I've never worked with it in detail. I reminds me of the power of Oracle's syntax without the huge cost that only large companies can afford.
I think that explains the jobs. Then you have the people who make suboptimal choices, such as buying Oracle today, because nobody got fired for doing that.
my company uses postgresql and dotnet efcore and it works really really really well. npgsql/efcore.pg is really awesome and has stuff built-in that works. kudos to roji which basically built it (as far as I know does he work for microsoft to bring npgsql forward). but maybe it works equally well, because microsoft bought citus and thus many psql developers.
.NET divers work great. As for Oracle when companies have been around for decades and rely on it I can see it's almost impossible to change without a huge risk to the business. Bu t if someone were starting a business today Oracle doesn't feel the way to go in most cases. Of course I'm sure they have some really good sales people that make it happen lol
However, I also really like using clis when they are available. In that sense psql is quite good and pgcli is really cool.
Nothing beats JetBrains DataGrip for me. I’ve tried a whole bunch of alternatives, but it’s most well connected and useful in terms of functionality.
I've used JetBrains for Development before and consider their quality exception so next time I use PostgresSQL I'll give it a try.
I've also used Navicat for SQLite and like it so I'll probably try that one as well.
Why is it better than MySQL? This article wasted my time. Is there a better article which explains why Postgres is better than other free and opensource SQL implementations?
In general, Postgres also seems to have better plugins and extensions
And yet lacks the most important one. A natively supported clustering and HA solution.
Especially important with CitusData being acquired by Microsoft and its future now at risk.
b) Is not part of the core PostgreSQL codebase and so you are forever having to worry about compatibility.
c) PostgreSQL still lacks multi-master HA which is needed when you want to HA but continue to scale out.
An actively developed FOSS solution backed by EnterpriseDB also one of the main company supporting PostgreSQL.
But err... And ahem ahem... there are more https://www.postgresql.org/download/products/3-clusteringrep...
It's a choice of the PostgreSQL development team to only include low level api for HA in core project. But it's fairly harsh to say there is no active support for HA. Many maintainers of the core are also invested in HA related project build around PostgreSQL.
Dealing with third party solutions means having to worry about incompatibility, enterprise support, ongoing maintenance etc.
There are a lot of companies out there which provide enterprise support for PostgreSQL + HA.
I have to work with oracle for a client right now and I’m gobsmacked that they don’t offer it at the pricepoint.
Every busy production db I've ever seen has had big problems when dealing with 10-hour transactions, and I don't see how Postgres could deal with these kind of problems as they're fundamental in the data model.
> Adding a column with a volatile DEFAULT or changing the type of an existing column will require the entire table and its indexes to be rewritten. As an exception, when changing the type of an existing column, if the USING clause does not change the column contents and the old type is either binary coercible to the new type or an unconstrained domain over the new type, a table rewrite is not needed; but any indexes on the affected columns must still be rebuilt. Table and/or index rebuilds may take a significant amount of time for a large table; and will temporarily require as much as double the disk space.
Clearly there are workarounds where you make a new empty column that you then slowly backfill with the correct value and eventually transactionally rename the new column to the old one, but the magic "oh postgres does DDL in a transaction so it's magic and we never have to worry" is pretty much gone at that point.
It's not a redo/undo log. It's more like (simplifying things here) having multiple copies of things and different sessions/connections "see" different sets of data. That's why you need to vacuum the database periodically, to garbage collect old rows that are no long accessible to any sessions/transactions.
See https://devcenter.heroku.com/articles/postgresql-concurrency
>is there enough storage space for that?
You do need to make sure you aren't too close to filling your storage, postgres needs some breathing room for this reason.
You'd just run your migration statements one after the other and pray it doesn't fail midway? Or have a contingency plan for every possible point of failure?
Postgres has given me a sheltered life.
If you value more expressive data types and constructs to read/write your data then PostgreSQL is better. If you don't need those but do need to be able to scale to very large data volumes then MySQL is better.
I don't understand why people keep wanting to make this a competition. It's been going for decades and it was stupid back then as it is today. The databases are both popular, both high quality but serve different needs.
https://wiki.postgresql.org/wiki/Foreign_data_wrappers IMAP, GitHub, BigQuery, Faker and CSV have been awesome for me, but don’t tend to be 100% supported by public cloud managed DB services. I’ve heard some can be flaky.
INSERT INTO public.people SELECT ssn, name, phone_number FROM faker.people;
That is a way to generate a ton of fake records for integration and load testing. Very elegant!
As a language it’s a bit clunky* but once you learn it, and you start moving database logic into the database, suddenly everything is 25% LOC, often literally 100x faster, and easier to understand.
As an added bonus, when I’ve moved very complex data transforms into pl/pgsql, the non-programmer technical people can now read the logic and have more than once pointed out problems (with solutions).
* it helps a lot if you write your SELECT and other statements in lowercase. Feels barbaric I know, but reading shoutycode all day gives me a headache.
At first the Oracle version felt clunky to me too but then after taking time to learn it I also found moving complex app logic to SQL allowed me to make complex app changes much quicker. I have use that method of development now for years in a variety of apps and databases and it works great.
Cross fingers now that my hubris does not come and bite me.
2. PostgreSQL Query Optimization: The Ultimate Guide to Building Efficient Queries - Henrietta Dombrovskaya, Boris Novikov, Anna Bailliekova is quite new, and exclusively focuses on query side of things, but goes deep into this topic. I haven't read it though, only flipped through the pages but it seems good.
And so, I find it really difficult to justify investing time in building up infrastructure to support it for my clients' work, because I need to be effective from day one.
Realistically that has absolutely nothing to do with PostgreSQL but whether or not some little third-party company or one or two individual developers have built something solid like Sequel Ace. That's really troubling for me to think about, but these CLI tools are insufficient for being productive.
This a great example to me as to why a business can't use a potentially superior product because some ecosystem feature is non-existent, or is hard to find.
That being said, I'm not too upset because modern MySQL seems perfectly fine, and still more widely used than PostgreSQL.
* Postico 1 is great, but the soon-to-be released Postico 2 is in beta atm and even better.
Wondering whether this is a missed opportunity. This is not to bash the free software part of PostgreSQL, but rather to focus on something that could drive more enterprise adoption, I guess to the detriment of primarily Oracle and Microsoft SQL Server.
"After 25 years of persistence, and a better logo design, Postgres stands today as one of the most significant pillars of profound achievement in free software, alongside the likes of Linux and Firefox." The better logo really does it. All three of the "pillars" the author names are clones or derivatives of commercial products (Ingres, Unix, Netscape Navigator). Postgres development was partly funded by DARPA at Berkeley, and then through licensing to IBM, before it was open-sourced. Only Linux can be called a "profound achievement" of free software if the author means the cathedral and bazaar model, and then just barely.
"PostgreSQL has taken a complex problem and solved it to such an effective degree that all of its competitors are essentially obsolete, perhaps with the exception of SQLite." Uh, no. The whole notion of "competitors" in the world of open source software is flawed. PostgreSQL has not made Oracle, SQL Server, and DB/2 obsolete, nor has it made MySQL obsolete, or any of the other database managers we have to choose from. PostgreSQL is about half as "popular" as MySQL (https://db-engines.com/en/ranking), and both are below Oracle.
"... providing the best implementation of SQL" and in the footnote "No qualifiers. It’s straight-up the best implementation of SQL." I can't tell what the author means by "best," it's like writing that Oreos are the "best" cookies. In terms of standards compliance DB/2 is better. In terms of performance and scalability Oracle, SQL Server, and DB/2 are all objectively "better." In terms of compared to the main open source alternative, MySQL, the two are nearly neck-and-neck with features, compliance, and performance.
"SQL is usually the #1 bottleneck in web applications, and Postgres does an excellent job of providing you with the tools necessary to manage that bottleneck." Not in my experience. Sometimes excessive database traffic causes performance issues in web apps, but many other problems show up, like Apache/nginx contending for resources with MySQL or PostgreSQL (hint: don't run those on the same server/VPS). If the code makes too many database queries, or badly-written queries (WordPress and pretty much every ORM) that's a fault in the application code, not the database, and PostgreSQL doesn't magically fix it.
Postgres supports table partitioning perfectly fine[1].
1. https://www.postgresql.org/docs/current/ddl-partitioning.htm...
I do tend to get attached to the DB of my last project because I discover new cool things. And there are some differences.
But I cannot think of a PostgreSQL project that would not also have worked great in MySQL or a MySQL project that could not have been done with PostgreSQL.
I have re-implemented systems that used non open-opensource DBMS's on Windows and the convenience and performance (on the same hardware) were just amazing when using PostgreSQL or MariaDB.
Can MySQL do GIS related features like PostGIS? Also IIRC to index JSON in MySQL you need to add generated columns for all the properties/paths you want to index, which can get pretty hairy.
It can[1] for most of the common cases, but it lacks many of the more advanced features PostGIS has.
So if you have a MySQL database up and running, you probably don't have to switch to PostGIS just because you want to add some geospatial data. However if starting a new geospatial project then you should almost certainly choose PostGIS.
[1] https://dev.mysql.com/doc/refman/8.0/en/spatial-types.html
DELETE FROM jobs WHERE id = (
SELECT id FROM jobs FOR UPDATE SKIP LOCKED
) RETURNING *;
MariaDB 10.5+ supports a RETURNING clause, and 10.6 now has FOR UPDATE SKIP LOCKED. This is quite new though (10.6 went GA a month ago) so there might be some wrinkles.
[1] https://mysqlserverteam.com/mysql-8-0-1-using-skip-locked-an...