Come on: it has nothing to do with memory: a virtual particle pair can popup anywhere at random.
Yes, this is the conventional view.
However, i'm considering this from the pov of your model.
We do know that energy is strictly quantized, do you challenge that?
Therefore, if that's true, a "virtual particle" can not be fractional, it can only happen when the variance exceeds a quantum.
The point being, a copy-and-delete operation has to preserve the original symmetries.