Hi Rafael, Building on the reply from Marco, we have experienced the same issue with queues and are evaluating a multi kahadb set up of a kahadb per queue.
Obviously this does increase the total number of files but so far we have not seen any significant increase in total storage compared to a single kahadb, and no change in the performance. It helps when there are old msgs because the journals are distinct and therefore journal files on *other* queues can be released without waiting for the old msg to be consumed. We haven't found a better way, but would love to know if there is one. Testing is still in progress before we move this configuration into production. Regards, Mark Johnson Principal Product Architect
