Python slow

Content:

Original link: Python Is Too Slow for Real Applications (And I Can Prove It) / CodexLab.

I am sure that the article is true.

But it is sort of saying that water is wet.

Everybody know that:

• running compiled code is faster than interpreting code
• multi-threading does not scale for interpreting if a GIL is present

Python is interpreted and has a GIL (actually recent versions has allowed free-threading aka no GIL and has an experimental JIT compiler, but traditional). Go and Java are compiled (AOT and JIT respectively).

So if one has a compute intensive task that needs to be done in parallel, then Go and Java are faster than Python. No surprise.

But I do not consider that a problem for Python. I consider that a case of choosing the right tool for the job.

Comments: