Robin,

>I have a coleague working in the area of CAD, who is interested in
>what kind of difference it would make if a CAD run took 2 hours, or 15
>minutes, instead of the current 8-12 hours.  He thought (and I agree)
>that there ought to be a parallel to how long it takes to compile a
>program.  They are in the equivalent of the era where we submitted a
>card deck and waited a day for turn around.

In the 'old' days factors other than cpu time were a significant portion of
the elapsed time needed to get a job done.

These days there is no connection between the time taken to compile a
program and the time taken to execute it.

>   Does anyone know of studies about what changes when compiles
>drastically change the amount of time they take?

Yes.  The main issues are number of bytes of source code (these days
these are mostly contained within header files) and the degree of compiler
optimization that has been requested.

I am not sure whether you are asking how to increase the speed of execution
of the CAD program, or increase the rate it is compiled at.  Your timing 
information
suggests the former.

In the former case you have three options:

    1) Compile the source at a greater optimization level (assumes you have 
source).

    2) Change hardware platforms.  CAD programs are floating-point intensive.
        I see you work for Sun, who make good web servers, but their 
floating-point
        performance is not exactly stunning.  You colleague might like to 
check out SGI
        or HP.

    3) Parallel processing.  Much talked about, but rarely achieved in 
practice.

If it is the latter case I would suggest looking at the include 
files.  Over time source files
accumulate includes that they no longer need.  CAD, so the source could be 
Fortran,
off hand I cannot think of any include checking tools for Fortran on Sun.



derek

--
Derek M Jones                                              tel: +44 (0) 
1252 520 667
Knowledge Software 
Ltd                                 mailto:[EMAIL PROTECTED]
Applications Standards Conformance Testing   http://www.knosof.co.uk



- Automatic footer for [EMAIL PROTECTED] ----------------------------------
To unsubscribe from this list, mail [EMAIL PROTECTED]  unsubscribe discuss
To join the announcements list, mail [EMAIL PROTECTED] subscribe announce
To receive a help file, mail [EMAIL PROTECTED]         help
This list is archived at http://www.mail-archive.com/discuss%40ppig.org/
If you have any problems or questions, please mail [EMAIL PROTECTED]

Reply via email to