I have made some progress with this. Switching from InMemory to DHT
has resolved the CPU issue (task manager) issue. now the box is
hovering between 4% and 12% instead of 100%.  Memory Usage (task
manager) still climbs but at a much slower pace. it could be the
number of messages that are flowing through the system.

I probably need to run my service through .dotTrace for memory
profiling. However I'm not really sure what to look for or when
profile setting to select.

On Aug 7, 10:38 am, Jason Meckley <[email protected]> wrote:
> My windows service seems to constantly increase is memory consumption
> (task manager>processes>memory usage). My first thought is this has to
> do with Saga State.
>
> Currently I'm using the InMemorySagaStatePersister. I thought this
> might be the issue, maybe in memory is for testing, not production
> (like NH has a HashTable cache for testing). Is this correct?
>
> If the memory consumption is due to the InMemorySagaStatePersister I
> have a choice between DistributedHashTableSagaSatePersister and
> OptimisticDistributedHashTableSagaSatePersister. What is the
> difference?
>
> Implementing either DHT or ODHT would require my state objects to
> Implement IVersionedState. This is simple enough with auto properties.
> I don't have the set the versioning properties, RSB handles that,
> correct?
>
> I then need to register mergers for my State objects. Again that is
> simple enough; implementing ISagaStateMerger<>. When I create the
> resulting state I don't need to worry about the versioned properties,
> correct? just set the properties that contain business value and
> return. RSB will handle version properites.
--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups 
"Rhino Tools Dev" group.
To post to this group, send email to [email protected]
To unsubscribe from this group, send email to 
[email protected]
For more options, visit this group at 
http://groups.google.com/group/rhino-tools-dev?hl=en
-~----------~----~----~----~------~----~------~--~---

Reply via email to