> This is very interesting as it is a technique I currently use, > any idea why there is such a performance hit?
Well, yeah... any time you are copying a whole bunch of "stuff" from one place to another, even just in RAM, there's gonna be a hit. The more "stuff" the bigger the hit. Pretty basic, and true of any technology, not unique to CF. In my case, I think we probably are storing about 4-8 megs of data in the app scope. Copying that into the request scope * 5 simultaneous requests = 20 to 40 megs or data constantly getting shuffled around, created, copied, destroyed, etc. I'm actually surprised the performance hit wasn't more than it was... -Cameron -------------------- Cameron Childress elliptIQ Inc. p.770.460.1035.232 f.770.460.0963 -- http://www.neighborware.com America's Leading Community Network Software > -----Original Message----- > From: Kola Oyedeji [mailto:[EMAIL PROTECTED]] > Sent: Saturday, December 29, 2001 9:21 AM > To: CF-Talk > Subject: RE: Session variables > > > This is very interesting as it is a technique I currently use, > any idea why > there > is such a performance hit? > > Kola > > -----Original Message----- > From: Cameron Childress [mailto:[EMAIL PROTECTED]] > Sent: 27 December 2001 21:37 > To: CF-Talk > Subject: RE: Session variables > > > > I would presume, however, that if you have a handful of simple variables > > like for instance a datasource name and a couple of primary keys that it > > should't have problems, correct? I would think the problem would > > come in if > > you are transfering queries, arrays of structs. etc. or am I just > > optimistic? I have always used the copy to request scope for frequently > > referenced variables not ocasional ones. It doesnt make sense to > > copy a huge > > struct which is seldom referenced, on every page request (can anyone say > > Spectra?). > > I would expect the performance hit to decrease as the > quantity/size/complexity of the information being copied decreased. > However, we have switched entirely away from the copy method and did not > test any point in between "copy insane amount of data" and "no > copying, lock > everything". > > The case in my example was a "worst case" because of the large amount of > data we had been storing in the Application scope, but it's a > great example > that a slowdown does actually exist and is not just theoretical. > > -Cameron > > -------------------- > Cameron Childress > elliptIQ Inc. > p.770.460.1035.232 > f.770.460.0963 > -- > http://www.neighborware.com > America's Leading Community Network Software > > ______________________________________________________________________ Dedicated Windows 2000 Server PIII 800 / 256 MB RAM / 40 GB HD / 20 GB MO/XFER Instant Activation � $99/Month � Free Setup http://www.pennyhost.com/redirect.cfm?adcode=coldfusiona FAQ: http://www.thenetprofits.co.uk/coldfusion/faq Archives: http://www.mail-archive.com/[email protected]/ Unsubscribe: http://www.houseoffusion.com/index.cfm?sidebar=lists
