Seems like you're over complicating it, use fork directly or Parallel::ForkManager, then use Storable to retreive the data from disk (or ramdisk I suppose). Since you're on Windows, the following may be more relevant (based on my quick bing'ing):

• Fork::Super
• snack overblow on fork on Windows

The benefit to using a fork based approach is you can control the parent program to make it async unti you don't want it to be by using wait.

Final note, using something like redis if it is available really makes it easy to share data among processes. This is what's it is for. Or maybe sqlite.