The goal of the Alliance project is to create a stable operating system based on a modified version of the Stanford Caching Model of Operating System Functionality. The primary modification makes it usable on Intel compatible hardware (the Stanford kernel was implemented on a PaRadiGM system built for that purpose). Alliance introduces a modern system for distributed computing, with a CORBA ORB as a communication backbone for the external kernels.
In the caching model as much functionality is transferred from the low-level Cache Kernel to middle level Application Kernels. The Application Kernels provide medium-level services to applications running on top of them. Basically, the only things the cache kernel manages are the creation of kernels, threads, and the address spaces they run in, as well as the multitasking of the threads. This design provides a suitable environment for the emulation of other operating systems.
The Cache Kernel’s design appeal comes from having the modularity of microkernels with the speed of monolithic kernels. This modularity combined with speed offers the possibility of OS emulation at acceptable speeds. It is not, however, the primary purpose of Alliance to provide emulations for all the existing platforms. This would obviously be impossible. Instead, we are looking to provide an environment to make emulation possible and relatively easy to implement.
To provide a platform to test the design and provide support for binaries and source from other environments for people switching to Alliance, we plan to convert one or two existing free operating systems (Linux and FreeDOS) to Alliance AK’s. This will give users of existing operating systems a familiar way to use a new kernel design.
Alliance is in no way tied to Stanford University. We are simply basing our work on their model. A modern model for OS design which avoids the inflexible monolithic kernel and slow and memory-hungry microkernel.
Earlier, the Alliance team used a closed development model to develop the initial specifications and code for the core system. This model has now been replaced by an open model.
We would like to welcome and encourage everybody to join the Alliance project. In the open model, this simply involves joining the mailing list of the subsystem you would like to work on, and discussing development with others. We also ask people to join the and supplying him with a username and password. Please direct any questions you have about membership or the Member Directory to .
The new open model is essentially the same as the development model used in other open development projects, such as the Linux operating system. The coordination of addition or change of code in the ‘official’ codebase is done for each subsystem by its original team.