Front page | perl.beginners |
Postings from August 2009
Hi Luke, luke devon wrote: > I wanted to delete some unwanted data in one of my oracle > database.20 000 of records are there. This is actually on production . > manually to delete a record , it takes 2 mins of time. > > So now i decided to write a perl script to do this function > by learning on the web. Can somebody give me a exact guide to delete > that amount of data , without giving any impact for performance of the > database and i would like to ask what is the better way to do such a > execution ? > Do we have to implement more threads ? Perhaps I'm misunderstanding the problem, but if you would like to remove 20,000 records in an Oracle database, I don't believe that Perl will help you with that. It seems that you're already doing everything in a single SQL statement; having Perl do it won't give you much added benefit (i.e., no matter how you do it, you still have 20,000 records to delete). If these records don't overlap, then you could consider a multi-threaded approach...I guess. Another suggestion if this is a production server is to mark records as stale rather than deleting them (i.e., add a boolean flag as another column). And then, when the server is least busy, delete the marked records. Or, rather than a boolean flag, assign some number (say, from 0 to 9) and periodically delete a portion of the rows...sort of like how log files are handled [delete the oldest ones]. I presume your problem is that deleting 20,000 in one go is slowing down the Oracle server? RayThread Previous | Thread Next