Saying hello.

Been thinking about distributed design patterns recently.
I have been following several technology stacks recently.

IMDBS - In memory database management systems.

Basically a means of distributing highly available near L2 memory access. I'm looking to use this in a current design pattern for high access to volatile working memory sets.  I have business logic that needs to be processed in a high speed manner.  I'm going to be using an in memory data store for my business logic storage.

Process A - fetches in memory store data...
  - acts upon data
  - pushes the end result back up into the in memory store
  - updates routing slip activity information ( state engine)
  - finishes process A publishes message w/ in memory store key information. and routing slip.

Process B - fetches in memory store data...
 - acts upon data
 - pushes the end result into memory store process b.

Process C - etc...

Process End
- acts upon data
- completes processing
- publishes cleanup message

Cleanup Message
 - removes Process A-N processing temporal data
  (possibly conduct this in line potential side affects with message is dropped)