Hi, We are in a process of building a CMS on top of Jackrabbit which exposes all the functionalities through REST. The frontend will be in PHP. Currently, the repository is deployed using Deployment Model 1 with the FileSystem as backend.
I wanted to know how can we leverage Clustering feature of JackRabbit for high scalibility. From reading the mailing lists and the Clustering Doc[1] on Wiki I deduced that it can be done only using a Database as a backend. My only concern is the storage of large files, which in our case can be Videos, Images, Audio and Text in DB as BLOBs. I have the following two queries regarding this. 1. Does the repository with huge amount of binary content gives comparable performance to Filesystem when DB is used as a backend? It'll be a great help if there are some figures for the performance of the Repository with DB. 2. Can clustering be done using externalBLOBs=true with everthing else being stored in DB? Also please point me to any relevant docs/mails in case there are any for me to read up. We will use MySQL as a DB in case a need arises. regards Vibhu [1] http://wiki.apache.org/jackrabbit/Clustering
