Distributing Subsystems Across Different Kernels Running Simultaneously in a Multi-Core Architecture

In current scenario, most operating systems are designed towards serving only one of the classes of computing i.e. Server, Interactive and Real Time and try to do justice to the other two classes to a reasonable extent. We propose a technique to overcome the limitation when a single hardware box is required to fulfill multiple computing classes such as Server, Interactive and Real-Time by running different kernels simultaneously on different cores of a multi-core system and provide synchronization between the kernels using IPIs and shared memory. On division of the subsystems across kernels, the provided separate environment will enable the users to run multiple operating systems, each one the best for its class of computing. We are developing a prototype which divides the devices among the cores of a dual core Intel x86 machine. We are proposing a heterogeneous operating system environment on homogeneous hardware by identifying different classes of computing for a particular operating system. This idea finds applications in - Filers, Graphics Processing Engines, Intelligent Switches. Thus, by introducing parallelism amongst the subsystems, we propose to improve the overall efficiency and scalability of the system. Hence, we conclude that, this approach would enable users to run applications that require different operating system environments and would provide a separation of multiple environments for users.