Making Revolution faster with really big arrays

[Wouter]

On 28 Apr 2005, at 17:59, Dennis Brown wrote:


> Frank,
>
> This is a simplified algorithm that I think would best translate 
> directly between Transcript and PostgreSQL for a time trial.  Just 
> make sure the PostgreSQL code is an item by item algorithm also.
>
> Thanks,
> Dennis
>
> Transcript output on my machine:  1,000,000 Element Sum = 500500000 
> Elapsed Time = 11.467659 seconds.
>
> on mouseUp
>   global gTestArray
>   if number of lines of the keys of gTestArray is not 1000000 then 
> --already done the init
>     put empty into gTestArray --start with an empty a variable
>     repeat with x = 1 to 1000
>       repeat with y = 1 to 1000
>         put x into gTestArray[x,y]
>       end repeat
>     end repeat
>   end if
>   --
>   put the long seconds into st
>   Method6
>   get the long seconds
>   put " Elapsed Time ="&&(it - st)&&"seconds." after msg
> end mouseUp
>
> on Method6 -- Add elements using keyed arrays
>   global gTestArray
>   put 0 into total
>   repeat with x = 1 to 1000
>     repeat with y = 1 to 1000
>       add gTestArray[x,y] to total
>     end repeat
>   end repeat
>   put "1,000,000 Element Sum = "&total
> end Method6
> -- 
>

Hi again with a little addition,

The initialization of the test array takes about 36 seconds.
Following version does the same in much shorter time:

on mouseUp
   put the long seconds into zap
   --if number of lines of the keys of gTestArray is not 1000000 then 
--already done the init
   put empty into gTestArray --start with an empty a variable
   repeat with i = 1 to 1000
     put i&comma after x
   end repeat
   repeat 1000
     put x after gTestArray
   end repeat
   split gTestArray by comma
   put the long seconds - zap
   --end if
   put the long seconds into st
   Method6
   get the long seconds
   put " Elapsed Time ="&&(it - st)&&"seconds." after msg
end mouseUp

In this case the initialization takes about 10 seconds on the slowbook