> At the time Tcl was the only multi-threaded GIL free scripting language for database access and this still holds true today.
Raku (nee Perl 6) is GIL-free IIRC. So are Julia and Erlang/Elixir, for sufficiently-loose definitions of "scripting language". The list probably goes on.
The tcl threads are shared nothing with message passing...each thread is a separate interpreter. So it's not obviously better or different than any other language running multiple processes with message passing via mmap or similar.
I feel that this statement is true for most values of X and Y. I've yet to find a mainstream language that's slower than python, or a task at which python is truly the most performant.
Fun fact: because of tkinter, Python embeds a TCL interpreter:
>>> import tkinter as tk
>>> tk.Tcl().eval('puts {hello, world}')
hello, world
''
So you could technically use TCL to query your DB if you cared.
But honestly, DB queries in real life is spread on threads, but then multiple processes, then a connection pool like pg bouncer, so you will unlikely see the difference most of the time. Not to mention the DB choice, the schema, the query quality, the DB configuration, the hardware and how complicated your request are all going to be more likely to affect the speed of request.
Still, interesting. Exploiting all those CPU in Python would cost a lot of RAM given how many processes you would have to spawn.
I would love to see the comparison using asyncio on uvloop to check the CPU are really the bottleneck though.
I'd say right in the early 2000s and then Django starts becoming popular, and the rest is history. But the writing is really on the wall in the late 1990s due to the intuitiveness of the Python language and beginner friendly design compared to the other popular scripting languages at the time including Perl, TCL, PHP, Ruby, etc. Speed, however, is never Python's advantage and for scripting languages it's not really a factor.
I think based on Python background it's safe to say that in order to become a hugely popular programming language, any language need to be beginner friendly and/or need a killer application as a motivation for programmers to adopt it. Speed and performance are not important factors unless you are writing an OS and we have already C for that.
A nice little thing that TCL has "given" Python, so to speak, is tkinter. Absolutely basic, but it still did the job for me back in the day.
I remember that when I started learning Python about 20 years ago I used to write a Python script with a quick and dirty tkinter interface on my Linux desktop at home, I was then copying everything on a floppy-disk in order to use that script/program at work, on a Windows machine (with the help of py2exe, another very nice tool).
Haha yeah ok you're right, but I like how he wrote it up in a nice and methodical manner, and the post sort of gives a sense that he enjoyed doing it and writing it up, which transfers. It's a peculiar thing that benchmarking has, I think, it becomes enjoyable in itself.
Meh, mp is kind of a disgusting hack. Maybe it would have worked here but it's so easy to run into bullshit. Off the top of my head, if you want to share a TLS connection you're obviously fucked, if you use gRPC you need to specify an env flag to be fork safe, memory overhead that interacts poorly with reference counted GC, etc.
Again, in this example I suspect things would have worked, because it's trivial code, but I wouldn't actually bet on it because I've seen MP fuck up so many times historically. Like, maybe one of those modules actually creates a static connection pool that gets messed up on mp? Who knows?
So yeah, maybe the obscene hack that 'mp' is would have worked or just use tcl, which is sane and worked for them. A note about mp in the post would have made sense, I just won't blame someone for not wanting to deal with it.
> if you want to share a TLS connection you're obviously fucked...
Some would argue that if you're using threads, well then you're fucked from the start, very possibly including me.
Go gets around this using go-routines, but go isn't necessarily thread safe per se. Futher go also has it's own bag of bugs you have to be aware of when using it.
Not quite, subinterpreters can contain multiple threads each. But you're right in that if you limit yourself to one thread per subinterpreter, then you should be able to get good multicore scaling that way, when more fully fleshed out subinterpreters arrive in 3.12. The trick then becomes how to efficiently communicate between them. Interesting times ahead.
You mean multiprocessing where communication goes either through serialization or through shared-memory where you have to use C-style programming and you lose all the advantages of e.g. a garbage collector and Pythonic datatypes?
Yeah the article felt intellectually honest and I enjoyed it, but was hoping for at least a footnote that multithreading is 1 solution. Async/await is another possible solution. There appears to be no shared state. But in fairness, python's state of multiprocessing, multithreading, and async/await just leaves too many options for a general language programmer to keep track of and perhaps it's just lack of knowing these things exist.
> (And although the Python test script could run in multiple processes HammerDB is an application simulating multiple virtual users in a GUI and CLI environment requiring the close interaction and control of using multiple threads).
