in reply to Re: Output isn't complete
in thread Output isn't complete

I should have mentioned that I am very new to Perl. I am not running this on anything larger than 250gb. The speed problem is likely due to my inexperience in programming with Perl.

If you don't mind my asking, how would I go about merging the find_subdirs together? If I could even cut the run-time on this script in half, I would be ecstatic.

Replies are listed 'Best First'.
Re: Re: Re: Output isn't complete
by waswas-fng (Curate) on Jul 11, 2003 at 16:19 UTC
    Just so I know for sure, could you give me a few lines from the input file so I know for sure what you are dealing with. After that I will have time this afternoon to refactor your code and see how much faster I can get it.

    -Waswas
      Here is an example of the input file:

      \\server1\D$\accounting\users
      \\server1\D$\finance\users
      \\server2\E$\corporate\users
      \\server2\E$\tech\users

      I really appreciate your help!
        Find A Share size in Win2k discusses getting the size of a share, but is applicable to any directory structure on Win32. Look at BrowserUK's solution using the FileSystemObject. FSO is a thin wrapper around the Win32 API calls. It should be quite a bit faster than File::Find.

        HTH

        I will take a step back and say that speeding this up may be hard, I thought you were doing this find on the local filesystem. Your slownedd issue is because you are doing this over the network (CIFS). I will take a moment and look through your code to see where the cutoff is heppening though.

        -Waswas