This is good news.

I was trying to add Exasol to ClickBench (https://github.com/ClickHouse/ClickBench/) since 2016, but it was not possible due to the limitations and the fact that it required using a custom virtual machine image.

Now we should try it again...

This is bad news: it is not usable:

> 5.3 Licensee may not disclose any benchmarking or results of evaluating the Software without Exasol´s prior written consent

This seems like a leftover from the old enterprise licenses. I will see if we can get that changed.

We'll be happy to be part of Clickbench. Reach out to me and we can work together to make it happen.

I would also like to have something like this, but for "vintage" links - something that looks like it was from the late 90s.

I use them in tests, just for fun: https://github.com/ClickHouse/ClickHouse/blob/master/tests/q...

There was a "shadyurl". The site itself seems to be long gone, but this'll give you some context: https://www.mikelacher.com/work/shady-url/

There's an example shadyurl link in here: https://news.ycombinator.com/item?id=14628529

Funnily enough the domains appear to have been bought up and are now genuinely shady.

GPU databases can run a small subset of production workloads in a narrow combination of conditions.

There are plenty of GPU databases out there: mapD/OmniSci/HeavyDB, AresDB, BlazingSQL, Kinetika, BrytlytDB, SQReam, Alenka, ... Some of them are very niche, and the others are not even usable.

The query tab looks quite complex with all these content shards: https://hackerbook.dosaygo.com/?view=query

I have a much simpler database: https://play.clickhouse.com/play?user=play#U0VMRUNUIHRpbWUsI...

Does your database also runs offline/locally in the browser? Seems to be the reason for the large number of shards.

You can run it locally, but it is a client-server architecture, which means that something has to run behind the browser.

The test does not look realistic: https://github.com/Cranot/grouped-simd-hashtable/blob/master...

Better to use a few distributions of keys from production-like datasets, e.g., from ClickBench. Most of them will be Zipfian and also have different temporal locality.

Here is my list of good technology blogs: https://clickhouse.com/blog/tech-blogs

There are many more exposed MySQLs than MongoDBs:

https://www.shodan.io/search?query=mongodb https://www.shodan.io/search?query=mysql https://www.shodan.io/search?query=postgresql

But this must be proportional to the overall popularity.

ClickHouse can do it. Examples:

    https://play.clickhouse.com/

    clickhouse-client --host play.clickhouse.com --user play --secure

    ssh play.clickhouse.com

Yes but CH is not SQL.

Yes, SQL is a query language and clickhouse is a database that uses SQL as a query language, but I don't see why that's relevant.

Let's add Spice to ClickBench: https://benchmark.clickhouse.com/

I've implemented the same idea a few years ago: https://pastila.nl/

It uses DB at the backend.