Hey all, trying to give it a second spin to the GWC Integration proposal work items found Andreas' totally right complain about <http://projects.opengeo.org/CAPRA/ticket/610>. Just commented on it. I know everyone is busy to death but if you have a chance to go through the tickets that will affect you please try to adapt the estimates and content as appropriate.
Gabriel. On Thu, 2010-07-15 at 12:53 -0300, Gabriel Roldan wrote: > Hello all, > > Having to prioritize other tasks I'm giving the GWC Integration proposal > the status of "finished under protest" :) > Meaning it'd be great for any/all of you to give it a read and comment > on any issue/concern, at your earliest convenience, no rush. > We'll be back when time permits. > > <http://projects.opengeo.org/CAPRA/wiki/GWC_Integration_Proposal> > > Cheers, > Gabriel > > On 6/30/10 10:34 AM, Gabriel Roldan wrote: > > On 6/30/10 10:14 AM, Chris Holmes wrote: > >> > >>>>> Gabriel is about > >>>>> to commit a Least Recently Used cache to GWC, which will allow an admin > >>>>> to set a total max for the cache. > >>> Right now the diskquota is an opt-in process meaning there's no global > >>> cache size cap, but you need to set the limit on a layer by layer basis. > >>> I think it would be easy to add a global limit so any non explicitly > >>> configured layer gets evenly capped to cope up with the global limit. > >>> How does that sound? > >>> > >> > >> That's an ok measure, if the truly global limit is hard. But I think > >> it's a lot more likely that we'll have a few layers that get a ton of > >> access, and a number that just get viewed a couple times. It'd be much > >> better if the few that get accessed a ton have bigger caches at the > >> expense of those that get rarely accessed. > >> > >> But if that's hard then go ahead and build it the easy way, and we can > >> see the actual nature of use. > > > > Giving priority to the most used layers makes a lot of sense. > > What we could do is to configure the global limit to use LFU (Least > > Frequently Used) expiration policy instead of LRU, and to aggregate the > > stats from all layers when whipping out. I think that'd make the trick. > > > > Agreed on all the other comments too. > > > > Cheers, > > Gabriel > > -- Gabriel Roldan [email protected] Expert service straight from the developers
