Hacker News Reader

Seconded. I understand that Timescale extends Postgres, but I'd still like to see performance comparisons relative to other time-series databases such as InfluxDB or even TiDB.

Shameless self-plug: we're also building database technology, but for embedding vectors (https://milvus.io) rather than time-series data - when doing query performance comparisons, we get several orders of magnitude of performance improvement over traditional databases. It's an unfair comparison, so we generally avoid it.

I think it's fair to mention it, but with the specific caveat of "If your workload has these characteristics, we are a particularly good fit."

[Timescale co-founder]

We have published many, many benchmarks versus other database systems. IIRC all of them also made the front page of HackerNews.

Here are some of them, for your reading pleasure :-)

TimescaleDB vs. InfluxDB: https://www.timescale.com/blog/timescaledb-vs-influxdb-for-t...

TimescaleDB vs. ClickHouse: https://www.timescale.com/blog/what-is-clickhouse-how-does-i...

TimescaleDB vs. Timestream: https://www.timescale.com/blog/timescaledb-vs-amazon-timestr...

Cool. Now do VS PG+extensions like asked about up-thread.

Once I've read this, I never had to think twice on what to use.

https://valyala.medium.com/high-cardinality-tsdb-benchmarks-...

Same. I've been experimenting with the recently fully open sourced Citus 11 and it's pretty incredible. To the point that I have a hard time imagining NOT using it going forward.

Maybe I need to do my own comparison at some point.

(Timescale co-founder)

That's a fair question.

We find that most developers storing time-series data on Postgres are doing so without pg_partman. So we first wanted to provide a benchmark that would be useful to most developers.

This benchmark was also the result of months of dedicated work. So the team did spend a lot of time on this. Unfortunately, they ran out of time to cover pg_partman. But that comparison is on our list of things to do soon.

Thanks!

Most people storing time-series are also doing that outside of TimescaleDB.

So the relevant question isn't what a typical PostgreSQL user is doing, but whether someone wanting to optimize their storage should look at a PostgreSQL extension, or an upstart database like TimescaleDB.

NB: I work at Timescale.

TimescaleDB is a PostgreSQL extension, just to be clear.

Disclosure, I work at Timescale.

Though I didn't write this post, I'd imagine at least part of it is that it's already nearly 4000 words and a 15 minute read and we just didn't want to add another set of things to it, to be perfectly honest.

`pg_partman` is cool! I haven't used it in a while, and because it uses declarative partitioning, it has some locking issues that we address with our partitioning scheme, but implying that it is OSS and we're not in terms of things like data retention features is a bit misleading as well. The `drop_chunks` command used for data retention is in the Apache 2 licensed portion of Timescale.

But almost all of your posts and benchmarks are based on the closed source version of Timescale. Everywhere I have seen it is always recommend to use the closed source version to get decent performance out of it.

(Timescale co-founder)

Just to clarify: Nothing on Timescale is closed-source. It is all source available, all on Github. Some of it is Apache2 licensed, some of it is Timescale Licensed. And it is all free.

Since the Timescale License is not an open source license, it is a closed source license. You are right, it is source available, but source available is also closed source. It is closed because it is not open. And it might be free as in beer, but it is not free as in freedom.

The source code is available, when someone says something is closed source then it usually means that the source code is not publicly available.

Do you want amazing things? Everything can't be "free as in beer" wtf does that even mean, i don't get free beer from anywhere.

It is a phrase that Richard Stallman created, and well known in free software communities- https://www.wired.com/2006/09/free-as-in-beer/, https://www.gnu.org/philosophy/free-sw.html

"Open source" and "closed source" are not the only options. There are plenty of products out there where you're technically allowed to look at the source code, but very restricted in how you can legally use it. The "open" in "open source" is generally understood to mean that users have permission to use, modify and redistribute the software. (Without that permission, calling it "freeware", "shared source" or "source available" would be more accurate.)

That's what "free as in beer" means -- it's a well-established phrase meaning "zero monetary cost": https://en.wikipedia.org/wiki/Gratis_versus_libre

In the case of the non-Apache-licensed version of TimescaleDB, you're allowed to use the software without payment, and you can distribute unmodified copies. But you're essentially forbidden from letting users define their own schemas, or from modifying it or reusing components unless your modified version imposes that same restriction. (The exception is if you agree to transfer ownership of your changes back to Timescale.)

Nobody's saying that Timescale can't build a non-open-source database, only that they should be clear about which parts are actually open. In my opinion, describing it on the homepage as an "open-source relational database" and then promoting it by benchmarking the proprietary version is at least a little bit misleading.

My recollection is that the TS license simply has protection against using the TS code to compete with TS, ala Amazon RDS.

While some people on HN feel that this is an impurity they can’t live with, I personally think it’s a small price to pay to enable development of TS to continue. In my opinion, claiming that it’s closed source is somewhat dogmatic. Many open source licenses have some kind of restrictions on use; the GPL comes to mind.

I have always been more on the pragmatic side of the FOSS movement (open source versus the philosophical/moral stance of the FSF, but still have tremendous respect for the FSF). For pragmatic reasons I reject the Timescale License. It doesn't just prevent Amazon (that alone would be misguided enough, though) but also prevent anyone other than Timescale from hosting the software for me. That means even if Timescales current offerings lined up perfectly with my business, I would be locked into whatever decisions they make in the future which may not align. The chances for a successful community fork are greatly reduced under the restrictions of the Timescale license. It makes it impossible for the community to make contributions on equal footing to Timescale, thus anyone making contributions are just doing free work for a corporation, rather than contributing to a product the community benefits from just as much as the corporation.

Ah, I suppose you'd prefer they switched to Affero GPL v3 (which BTW IS open source and written by FSF itself) - check it out here:

https://www.gnu.org/licenses/agpl-3.0.en.html

This license turns out to be very difficult to use for almost developer.

I prefer the Apache or Mozilla Public Licenses myself. But I accept the AGPL and the GPL family of licenses will consider useing software licensed under them in limited ways (it isn't clear how the AGPL interacts with infrastructure software).

Sometimes when benchmarks are posted on HN that compares your own product (A) with a competitor's product (B), someone working for the competitor refutes the claims made in the blog and states that B were not given the same requirements as A. This particular thread comes to mind [1]. Maybe this is due to the blog writers misunderstanding B or maybe the competitor misunderstands A, or maybe there is a better way to do what was benchmarked using B. I think a better way would be to use some standard benchmark, and publish the code used for benchmarking them.

[1] https://news.ycombinator.com/item?id=31767858

Hello! I run Developer Advocacy at Timescale.

Easy fix - we can do a benchmark comparing TimescaleDB to pg_partman.

Longer reply: pg_partman does address many of the same developer experience items we do, but it doesn't offer things like compression or modified plans. It will perform roughly the same as declarative partitioning (because that's what it uses), and I'm guessing we will see results similar to the last large table (in the Declarative Partitioning section).