On: Async Python is not faster

[Spiritus]

Funny how in the TechEmpower Web Framework Benchmarks[1], the async frameworks are basically destroying their sync counterparts.

[1] https://www.techempower.com/benchmarks/

[calpaterson → Spiritus]

I think that is largely because they aren't using enough workers. IIRC they use 3 * cpu_count for sync, which just won't be enough.

They also misleadingly de-emphasise latency variation. One Python async framework I'd never heard of was top of the pops on throughput there even though the latency numbers suggested it had pretty much fallen apart in the test.

[hedora → Spiritus]

None of the performant async frameworks in those benchmarks are written in python (unless I missed one).

The takeaway of this article is that python’s async io implementations perform poorly.

That’s surprising, since async I/O is usually used for performance reasons, and in most other languages, async I/O can be much faster.

[Spiritus → hedora]

The idea is to filter so that it only shows Python. Not to compare with other languages.

[hu3 → Spiritus]

I'm suprised to see ASP.NET Core taking 1st and 3rd places in the plaintext benchmark.

https://www.techempower.com/benchmarks/#section=data-r19&hw=...