We have here a "non" profit project for which we use Alfresco to classify scanned pdf documents. Its about 2000 pdf files with a complete size of around 2GB. We use alfresco 2.1 community version on a Compaq Proliant ML 530 (Xeon 2.8Ghz, 2GB RAM) / Fedora 9.
We created around 100 Content rules to attach several categories to the files. After all reading of the avialable tuning documents and optimizing all underlying software components (Java, recompiled and optimized MySQL) the alfresco system is already REALLY slow. Just openting the list of the content rules takes around 1min.
Is there any other way (except of the jconsole) to profile the systems for finding the leak?
It would very help us because we are a little bit in time pressure.
100 rules seems rather excessive - are those all in the same space(s), or are each of those rules in their own unique space? If the former, is it feasible to merge all of the logic in these rules into a single Javascript script, and then configure it as a (single) "Execute a Script" rule in those space(s)?
Its for very special kind of recategorizing of complex document structures which isn' t the other way possible. The problem at Your suggestion is, that I don't have the knowledge to write webscripts/javascripts for alfresco. So I can program javascript/java etc. but I don't have the information about alfresco internal class structure. Can You attach me a short example how to attach a category to a file based on search term on its filename? This would be very helpful