Isolation API vs Process Model

Hi all,

Isolation API provides a uniform mechanism for managing Java application life cycles that are isolated from each other, at least with respect to object reachability, but that can potentially share underlying implementation resources -- which means it's only concern about memory safety: it guarantee isolation of Java state (i.e. logically disjoint heap spaces, per application mutable static fields, etc), and communication between isolates by links. But memory safety alone is not sufficient for an os: the system needs to support the separation of application and limit their use of resources, such as memory, cpu, and network bandwidth, (and of course inter-application communication).
Currently,it seems that method MemoryResourceImpl.claimMemoryResource(..) doesn't enforce memory usage policy to limit the available memory to a plugin, right? If so, the result would be catastrophic imaging a malicious driver instantiates 100,000 objects! K0 is a java system at the University of Utah, it support sepearte heap and garbage collection for each K0 process, thus limite the memory resource a process can consume. JNode is also multi-heap system, but it looks more like buddy memory allocation than memory management, right?
So, do we need a process model in JNode besides support Isolation -- as a process support protection, resources management and IPC, where isolation is only about protection.

Cheers

-Birkey