starrocks VS garnet

tidb has been around for a while, it is distributed, written in Go and Rust, and MySQL compatible. https://github.com/pingcap/tidb

Somewhat relatedly, StarRocks is also MySQL compatible, written in Java and C++, but it's tackling OLAP use-cases. https://github.com/StarRocks/starrocks

I think you're talking about doing denormalization before importing data into an OLAP system to avoid subsequent joins. However, this greatly limits the flexibility of data modeling. Moreover, denormalization can be a headache-inducing process. In fact, I have tested StarRocks (https://github.com/StarRocks/starrocks), and it is capable of performing joins while streaming data imports, and the speed is very fast. It's worth giving it a try.

https://github.com/apache/doris Is a great example. Same for it's cousin https://github.com/StarRocks/starrocks that was an early fork of the doris project.

To be fair, these are the only examples I can think of and I only learned of these as I'm standing up new data infra using starrocks.

ClickHouseClickHouse and Starrocks are similar. They are both columnar databases powered by vectorization tech, which means they are really fast.

StarRocks, it’s a Linux Foundation project now, but a lot of the initial team and community behind it came from China.

https://github.com/StarRocks/starrocks

Funny that I hadn’t heard of them in the database space till they showed up at the top of ClickBench. Makes me wonder what other interesting projects I’m missing out on in China.

Full disclosure - I work for StarRocks (starrocks.com)

First of all, this is great. Transparent and healthy competition is always great for the customers!

Regarding the joined table queries that are missing in the tests, this is exactly why we built StarRocks - to give people the best performance of complex analytics queries on both joined tables and single tables.

I encourage you to checkout this blog: https://starrocks.medium.com/starrocks-outperforms-clickhous...

And, give us a star if you think we are doing the right thing: https://github.com/StarRocks/starrocks

Follow us on LinkedIn for the latest updates: https://www.linkedin.com/company/starrocks

You would be surprised by performance of modern .NET :)

Writing no-alloc is oftentimes done by reducing complexity and not doing "stupid" tricks that actually work against JIT and CoreLib features.

For databases specifically, .NET is actually positioned very well with its low-level features (intrisics incl. SIMD, FFI, struct generics though not entirely low-level) and high-throughput GC.

Interesting example of this applied in practice is Garnet[0]/FASTER[1]. Keep in mind that its codebase still consist of un-idiomatic C# and you can do way better by further simplification, but it already does the job well enough.

[0] https://github.com/microsoft/garnet

[1] https://github.com/microsoft/FASTER

At the same time Microsoft releases Garnet: https://github.com/microsoft/garnet

Good timing.