you can do this yourself.

-setup 1..n JVMs. 
-Each JVM creates a drools instance with the same rules
-Create an facade interface where external application insert/update/retract 
are actually operations on an infinispan replication group.
-Each JVM attaches an insert/update/delete listener to the infinispan client, 
the listener translates the infinispan event into a working memory action. This 
allows all JVM's in the group to operate in redundancy.
-Make sure that any rules that have external side effects - like calling 
services - have an "if master" check, as you don't want the slave JVM's doing 
any external side effects. However you will need to make sure any return 
information from that side effect is replicate to all JVMS in the group. Either 
ISPN already has heart group management for master/slave, or you can use the 
underlying jgroups to do it.

If someone gets this working, they should probably publish it on github as an 
example project.

Mark
On 12 Nov 2012, at 15:29, groovenarula <gnaru...@la-z-boy.com> wrote:

> Follow up on high availability - 
> 
> I'm planning to use Drools for Capacity Management. In my use-case I need to
> track # of hours of planned capacity that has been consumed based on hours
> allocated. I'm planning on having a StatefulKnowledgeSession that will have
> facts that store the # of hours of capacity that are available for a given
> week. Then as work comes in (in the form of orders), it will be allocated
> against this capacity. Even though at a high level this seems to be a simple
> requirement, the process of allocating resources is very complicated and
> using rules and FOL would greatly help manage the complexity of the
> allocation process.
> 
> Based on the discussion above it looks like I won't be able to architect a
> solution that's highly available. I'm not concerned about performance (the
> drools engine is very efficient even on a single server). But it looks like
> it's not possible to share / replicate session state contained in a
> StatefulKnowledgeSession across separate JVMs (either on the same server or
> across 2 servers). 
> 
> Am I correct or am I missing something ?
> 
> 
> 
> --
> View this message in context: 
> http://drools.46999.n3.nabble.com/rules-users-drools-grid-or-other-high-availability-solution-tp4020734p4020803.html
> Sent from the Drools: User forum mailing list archive at Nabble.com.
> _______________________________________________
> rules-users mailing list
> rules-users@lists.jboss.org
> https://lists.jboss.org/mailman/listinfo/rules-users


_______________________________________________
rules-users mailing list
rules-users@lists.jboss.org
https://lists.jboss.org/mailman/listinfo/rules-users

Reply via email to