Hi, Is it possible to spread the messages to different queues with separate brokers and tip the balance decisively in the consumers favor.. ?
This is the only real way to accomplish what you need to do without compromising on QoS (memory, expiration etc) or on processing speed. All other models you are trying to put forth do not seem to adequately & uncompromisingly address the core issue. Cheers, Ashwin... Vomityn wrote: > > I have a rather unique problem, which I was hoping the Camel community > could assist me with. > > Take the following scenario: > A queue is being populated with messages regularly every few milliseconds. > The data within the message is only relevant for 5 seconds, at which point > it expires. > > The messages enter a volatile container in heap, which will be processed > every second. > > Any messages that have expired i.e older than 5 seconds will be removed > from the container. > > Once the process completes, a database will be updated with the latest > information. > > Given that roughly 600 messages will be processed per second, I'd like to > keep the container of relevant messages in heap. > </scenario> > > I've tried to implement an AggregatorStrategy to process the messages as > they come in, but storing messages for 5 seconds makes this difficult. > > I believe I can use Quartz to schedule the process each second, and along > with stateful=true, I can use the RAMJobStore. But I'm confused about how > to consume messages off the queue using a quartz endpoint. Is this where > pollEnrich comes in? > ----- --- Ashwin Karpe, Sr. Principal Consultant, PS - Opensource Center of Competence Progress Software Corporation 14 Oak Park Drive Bedford, MA 01730 --- +1-972-304-9084 (Office) +1-972-971-1700 (Mobile) ---- Blog: http://opensourceknowledge.blogspot.com/ -- View this message in context: http://old.nabble.com/Question-about-processing-and-persistence-tp28507720p28511783.html Sent from the Camel - Users mailing list archive at Nabble.com.
