Apparatus and method for a data processing system having a peer relationship among a plurality of central processing units

A data processing system is disclosed in which a plurality of central processing units have access to all the system resources, i.e., have a peer relationship. During initialization of the data processing system, all the system resources are allocated to the individual central processing units according to a preselected distribution, the identification of available resources thereafter being stored in the files of the individual central processing units. During the operation of the data processing system, the resources can be reallocated by a predetermined procedure. The central processing units entering such a relationship are required to include apparatus and/or software procedures that prevent access to system resources not assigned thereto. A mail box procedure, using locations in the main memory unit permit communication between the central processing units and are used in the dynamic allocation of resources.