Re: [Chicken-users] Hello World execution time

[Bobby Powers]

I got similar, but slightly different results (Fedora 14, 64-bit, core i7):

(time per run in wall seconds)

c: 0.001086

bash: 0.002742

py: 0.000336

scm: 0.005199

pl: 0.000344

so bash slightly beat out perl, and chicken came in last of the 5.  Didn't bother with mono or java.

Source available:

https://github.com/bobbyp/helloworlds

so on a *nix box:

$ git clone git://github.com/bobbyp/helloworlds.git

$ cd helloworlds

$ make

and it should give you a report shortly.

BUT, I think these tests lie.  I do the following:

$ strace -C -ttt ./hello.scm 2> chicken_report

$ strace -C -ttt ./hello.py 2> python_report

And strace nicely shows me for the chicken binary:

% time     seconds  usecs/call     calls    errors syscall

------ ----------- ----------- --------- --------- ----------------

100.00    0.000064           3        19           mmap

  0.00    0.000000           0         6           read

  0.00    0.000000           0         1           write

  0.00    0.000000           0        12         6 open

  0.00    0.000000           0         6           close

  0.00    0.000000           0         3         3 stat

  0.00    0.000000           0         8           fstat

  0.00    0.000000           0         1           lseek

  0.00    0.000000           0         8           mprotect

  0.00    0.000000           0         2           munmap

  0.00    0.000000           0         6           brk

  0.00    0.000000           0         1         1 access

  0.00    0.000000           0         1           execve

  0.00    0.000000           0         2           getrusage

  0.00    0.000000           0         1           arch_prctl

------ ----------- ----------- --------- --------- ----------------

100.00    0.000064                    77        10 total

So, only 0.000064 seconds spent in the kernel (Python had 0.000626, an order of magnitide more).  I think the reporting of execution time may be skewed by scheduling or something funky like that.  That, or python (and perl)'s runtime is a _lot_ faster to initialize than chicken. Is that possible?

yours,

Bobby

On Sun, Mar 13, 2011 at 8:26 PM, Daniel Aquino <address@hidden> wrote:

Here's one of the links: http://www.haskell.org/haskellwiki/Simple_unix_tools

On 3/13/11, Daniel Aquino <address@hidden> wrote:
> Haskell should be just as fast as c especially for something that
> basic.  There is a whole set of common unix utilities implemented in
> very short one liners.
>
> On 3/13/11, John J Foerch <address@hidden> wrote:
>> Tobia Conforto <address@hidden> writes:
>>> "Hello,
>>>
>>> World execution time"—may not be the most meaningful of benchmarks, but
>>> it's
>>> pretty important when you are writing shell scripts / cron jobs / random
>>> commandline utilities. It also serves to compare the startup overhead of
>>> different execution environments. So I ran this benchmark for my own
>>> curiosity
>>> and I thought you might like the results.
>>>
>>> Rules:
>>>
>>>   * the program should print "Hello, World!\n" and exit cleanly;
>>>   * no "benchmark modes" that would hinder real-world use of the
>>> language
>>> are
>>>     allowed;
>>>   * no -e allowed: each program should run from its own file (source,
>>> bytecode
>>>     or machine language as it may be.)
>>>
>>>
>>> I ran these on a fast, otherwise idle machine, doing 10 runs to warm it
>>> up, and
>>> then taking the median real time of 101 runs. (So yes, I like the median
>>> more
>>> than the mean, when measuring things.)
>>>
>>> [cid]
>>>
>>> The choice of languages is arbitrary. C is compiled, Mono and Java are
>>> poor-man's-compiled, the rest is interpreted. As for Chicken, don't
>>> bother
>>> asking: there is but a 2ms difference between csi and csc -O4 -block. I
>>> would
>>> have included Clojure, as I find the language itself not without its
>>> merits,
>>> but the current implementation is 4 times slower than plain Java and
>>> skewed the
>>> graph badly ;-)
>>>
>>> So that pretty much settles the question for me!
>>>
>>> cheers,
>>> Tobia
>>
>>
>> This is very interesting.  I would be interested to see haskell among
>> the set.
>>
>> --
>> John Foerch
>>
>>
>> _______________________________________________
>> Chicken-users mailing list
>> address@hidden
>> http://lists.nongnu.org/mailman/listinfo/chicken-users
>>
>
> --
> Sent from my mobile device
>

--
Sent from my mobile device

_______________________________________________
Chicken-users mailing list
address@hidden
http://lists.nongnu.org/mailman/listinfo/chicken-users