I suppose that what Matt was referring to, when talking about stateful CFCs scalability issue, is that you cannot use session replication if you put CFC in session (because for the moment, CFC are not serializable).
But, I don't think that it will prevent you to build scalable MVC applications built with CFC on the model layer... (most of the time, you would use sticky sessions with a load balancer in front of your cluster, without session replication). In 95% of CF applications, session replication capabilities is not a requirement. The best example is macromedia.com which extensively uses CFCs on the "model" layer (with View as ColdFusion pages or Flash applications). I suppose they also use CFC in session scope, don't you Sean? Benoit Hediard www.benorama.com > -----Message d'origine----- > De : [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] la > part de Matt Liotta > Envoye : mardi 25 mars 2003 14:25 > A : [EMAIL PROTECTED] > Objet : RE: [CFCDev] MVCF at benorama.com > > > > Certainly no problem with stateful CFCs. I think the problems to which > > you refer can mostly be traced to attempts to do one of two things (or > > possibly both at the same time): > > 1. trying to use cfproperty and "reflection" to describe persistent > > properties and then walking the CFC metadata to generate SQL. > > 2. trying to create generic base classes (mostly that do (1)) so that > > every instance created carries lots of "baggage" in terms of inherited > > methods and so on. > > > I agree that both of the above can lead to performance problems with > CFCs. However, neither is what I was referring to when I stated that > CFCs have scalability problems when persisted and/or when part of > complex inheritance hierarchies. > > Sean, you were one of the main participants in the threads where these > issues where discussed. Do you not recall the use cases where CFCs > didn't scale? > > -Matt > > ---------------------------------------------------------- > You are subscribed to cfcdev. To unsubscribe, send an email > to [EMAIL PROTECTED] with the word 'unsubscribe cfcdev' > in the message of the email. > > CFCDev is run by CFCZone (www.cfczone.org) and supported > by Mindtool, Corporation (www.mindtool.com). > > ---------------------------------------------------------- You are subscribed to cfcdev. To unsubscribe, send an email to [EMAIL PROTECTED] with the word 'unsubscribe cfcdev' in the message of the email. CFCDev is run by CFCZone (www.cfczone.org) and supported by Mindtool, Corporation (www.mindtool.com).
