sorry, but a disappointing performance report 
Author Message
 sorry, but a disappointing performance report

greetings,

here at my company they have been using a apl-like interpreter
in production mode for about a year.  we'll call it z language to
avoid proprietary issues.

we did a test in both languages:

   a=. i. 1e7
   +/a
4.999995e11

for the above, the z interpreter was an order of magnitude (10x) faster  
than the j interpreter.

the person who wrote it, a very smart person, says he didn't do anything
special or anything different than what a good c programmer would
noramlly do.   my quick inspection of his source code proves this to be  
true.

the z interpreter is an order of magnitude slower than compiled c code, so  
that means that j is 2 orders of magnitude (100x) slower than c.

this raises the following questions:

1.  can these results be verified by testing j against apl interpreters?

2.  can it be that j is doing additional work that z is not?

3.  does anyone really care about the speed?

4.  perhaps maybe someone should think about modifying to j interpreter
to have it run faster.

5.  i might be willing to take a look at it, if i could find the
documentation.  as it is i can't make head or tails out of the code.

joe.

==============================================================

!                                             (312) 732-3439 !
!                                                            !
! "The main thing is to keep the main thing the main thing." !
! Dr. George Sweeney                                         !
==============================================================



Sun, 25 Jun 1995 03:02:33 GMT  
 sorry, but a disappointing performance report

Quote:

>   a=. i. 1e7
>   +/a
>4.999995e11
>for the above, the z interpreter was an order of magnitude (10x) faster  
>than the j interpreter.

What release of j did you use? I understand it has been sped up of late.

Quote:
>the person who wrote it, a very smart person, says he didn't do anything
>special or anything different than what a good c programmer would
>noramlly do.   my quick inspection of his source code proves this to be  
>true.
>the z interpreter is an order of magnitude slower than compiled c code, so  
>that means that j is 2 orders of magnitude (100x) slower than c.
>this raises the following questions:
>1.  can these results be verified by testing j against apl interpreters?

There is a timing primitive in j. Try:(100) 6!:2 <'+/a' to get the time
average of 100 executions.

Quote:
>2.  can it be that j is doing additional work that z is not?
>3.  does anyone really care about the speed?
>4.  perhaps maybe someone should think about modifying to j interpreter
>to have it run faster.
>5.  i might be willing to take a look at it, if i could find the
>documentation.  as it is i can't make head or tails out of the code.

Documentation for the code is available for $90 from ISI in Canada.  It is
a terse, but IMHO, clear document. Also, Roger Hui has been awarded a
fellowship type of position by the ACM for the year 1993.  His travel to
sites in North America can be partially funded by the ACM.  He implemented
j and could probably motivate folks to enhance the code.

(B=)



Mon, 26 Jun 1995 04:48:33 GMT  
 
 [ 2 post ] 

 Relevant Pages 

1. Disappointing alpha performance

2. KVEWIN Changes and Monthend Report Program - Sorry

3. reporting a compiler performance "bug"

4. A little disappointed

5. C4b patch - Disappointed w/ Translator Class Fix

6. Disappointed with Eiffel

7. Disappointed with MF COBOL 4.0

8. Fujitsu -disappointing, RM-Cobol -primitive, anything else?

9. WAS: IVF continues to disappoint (relatively)

10. IVF continues to disappoint (relatively)

11. Still Disappointing libcpml.a for Alpha/Linux

 

 
Powered by phpBB® Forum Software