Hi,
I have a scenario where i have:
- A listener that listens to a port and collects datapackets(Logs of
Events) as they come and dumps to a DB.
- consider the data that goes into my table is parsed, processed
(meaningful, not raw anymore).
Can Drools sit in between this listener and my DataBase to achieve the
following :
1. Monitor the frequency of occurrence? Say if/when event 'X' occurs 'n'
times in a 'z' time window?
2. ignore event 'X' until it occurs n times.
3. track a sequence of events - event X followed by event y within a fixed
time window.
(the RHS for any of these is, at the moment, not very consequential. i might
just raise a flag or something to indicate something has happened...)
My query is primarily to understand if this can be achieved.
i may have around a million logs a day, but - only one event at a time is
passed from the listener to Drools. Can drools hold the knowledge of
previous facts(like count) into its working session? or does it require to
store all these facts also?
Should i be firing the rules whenever each log/event enters my system? or
should i accumulate the day's facts ? (but i might have rules that need to
catch conditions spreading over a few days ...)
Some pointers or suggestions on this would be great!
Thanks!
--
View this message in context:
http://www.nabble.com/Can-Drools-be-Used-for-this-application--tp23170278...
Sent from the drools - user mailing list archive at
Nabble.com.