In a multiprocessor system where a bank of processors share a bank of memories, it becomes necessary to equip a processor with the capability to periodically deny to other processors access to certain memory locations. We call this the memory-protection problem because the processor which has gained access to the data must protect it from others. The type of data contained in the memory locations to be protected dictates the characteristics of the protection techniques which are applicable. This paper outlines the techniques as they apply to four classes of data which are defined for multiprocessor systems. In those cases where special hardware is required, such hardware is described. Furthermore, the programming implications and the error control problems are discussed.

In multiprocessor systems it is desirable to look for and allocate storage without extensive data moving. Two techniques for accomplishing this, associative memory techniques and indirect addressing techniques, are described. It is concluded that the two methods are similar in performance, but that indirect addressing is more economical. An indirect addressing method is described in detail and various methods of implementation compared. The memory overhead and the time penalty (in memory accesses) are given as a function of memory size and method of implementation. It is concluded that a computed address table look-up technique should be used and that the block size should be either 256 or 512 words.